FAAOffice of Aviation Policy and Plans (APO-100)

FAA U.S. Passenger Airline Forecasts, Fiscal Years 2017-2037Methodology and Data Sources

March 24, 2017Version 1.0

Table of ContentsBackground................................................................................................................................................ 3Purpose of this document.......................................................................................................................... 4Document revision history......................................................................................................................... 4Acknowledgements.................................................................................................................................... 4Domestic forecast methodology................................................................................................................5

Forecast Years...................................................................................................................................... 5Assumptions.......................................................................................................................................... 5Domestic Forecast Methodology...........................................................................................................6Alternative Scenarios............................................................................................................................ 9

U.S. Airlines International Forecast............................................................................................................9Forecast Years.................................................................................................................................... 10Form 41 Forecast Methodology..........................................................................................................10Alternative Scenarios..........................................................................................................................12

U.S. and Foreign Flag International Forecast...........................................................................................12Forecast Years.................................................................................................................................... 13CBP Forecast Methodology................................................................................................................13

APPENDIX A: Glossary of terms.............................................................................................................18APPENDIX B: Data inputs and sources...................................................................................................19

Data inputs and sources for the baseline domestic forecast...............................................................19APPENDIX C: Model outputs....................................................................................................................27

Baseline Domestic Model Output........................................................................................................27Baseline International (Form 41) Model Output...................................................................................38Baseline International (Customs and Border Protection) Model Output..............................................42

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 2 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Background

The Federal Aviation Administration (FAA) Aerospace Forecast Report, henceforth referred to as the Report, is produced annually by the FAA’s Forecast and Performance Analysis Branch of the Office of Aviation Policy and Plans (APO-100). The Report covers the following subject areas:

U.S. airlines (passenger and cargo) General aviation U.S. commercial aircraft fleet Unmanned aircraft systems Commercial space transportation, and FAA operations at towers, Terminal Radar Approach Control and En-Route facilities

From this point onward, this document will only discuss the traffic and passenger forecasts developed for U.S. passenger airlines.

The Report details operations and passengers, over a twenty year period, for U.S. airlines flying domestically and internationally. These forecasts are used by the agency in its planning and decision-making processes. In addition, these forecasts are used extensively throughout the aviation and transportation communities as the industry plans for the future.

The forecasts can be found at this website: Link to Aviation Aerospace Forecas

In reading and using the information contained in the forecasts, it is important to recognize that forecasting is not an exact science. Forecast accuracy is largely dependent on underlying economic and political assumptions. While this always introduces some degree of uncertainty in the short-term, the long run average trends generally tend to be stable and accurate.

It should also be noted that the forecasts reflect unconstrained demand; that is, it is assumed that airports, air traffic control, and the airlines will increase supply as demand warrants.

Lastly, the forecasts represent only flights that enter or depart from the United States (U.S.) and do not include Unmanned Aerial Systems (UASs)1 nor low earth orbit flights.

1 Also known in the popular press as “drones.”FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 3 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Purpose of this document

The purpose of this document is to standardize the process, requirements, data sources and analyst judgment required to develop the national and international forecasts as well as provide a reference for anyone who uses them in their own analyses.

Updates to this document will be made on an on-going, as needed basis. Policy decisions, software updates, and data availability may necessitate changes. Any questions or comments should be directed to the individuals listed in the Acknowledgements section.

Document revision history

Revised by Roger Schaufele, APO-100 Date Revised March 21, 2017Revision Reason First draft Revision Control No. 1.0

Acknowledgements

This document was prepared by the FAA Forecast and Performance Analysis Branch of the Office of Aviation Policy and Plans under the direction of Roger Schaufele, Manager. The following individuals were responsible for individual subject areas:

Economic environment and general oversightRoger Schaufele, Manager202-267-3306Roger.Schaufele@FAA.gov

Domestic forecastsRoger Schaufele, Manager202-267-3306Roger.Schaufele@FAA.gov

International forecastsLi Ding202-267-1846Li.Ding@FAA.gov

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 4 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Domestic forecast methodology

Forecast YearsThe Report is published annually by the FAA and includes historical data and forecast data for a 20 year horizon. Historical and forecast data presented include:

Economic assumptions Available seat miles (ASMs) Revenue passenger miles (RPMs) Load factor (LF) Passenger miles flown Nominal and real passenger yield2

Enplaned passengers Average seats per aircraft mile Average passenger trip length (PTL) Forecast accuracy3

Alternative (optimistic and pessimistic) scenarios

Data in the Report are presented on a U.S. Government fiscal year basis (October through September). All model inputs are converted from calendar year to fiscal year when required.

AssumptionsThe Report assumes an unconstrained demand driven forecast for aviation services based upon national economic conditions as well as conditions within the aviation industry. It is “unconstrained” in the sense that over the long term, it is assumed that the aviation industry will expand (or contract) as necessary to meet demand. That said, it should be noted that some airports do function under constrained conditions (e.g., slot caps at LaGuardia airport) and that weather and unforeseen events like September 11, 2001 impact demand and the ability of the system to satisfy demand requirements in real time. These real world “constraints” are inherent in the historical data that the statistical models use to forecast the outputs bulleted above; therefore, they do influence the model’s “unconstrained” forecast.

Domestic Forecast MethodologyHistorical data used to supply inputs into the forecast models were obtained from U.S. Department of Transportation’s Bureau of Transportation Statistics. Additional information about the input data can be found in Appendix B.

2 Yield includes the following taxes and fees: FAA ad valorem tax, segment fee and Transportation Security Administration (TSA) security fee.3 The forecast accuracy section evaluates system totals, that is, the total of domestic and international forecast variables.FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 5 of 63Methodology and Data Sources Office of Aviation Policy and Plans

For statistical modeling, APO uses SAS software.4 To develop its short term (one year out) domestic and international forecasts of key traffic measures, the FAA uses a simplified version of the Unobserved Components Model (UCM)5 called the Basic Structural Model (BSM). The model is used to forecast enplaned passengers (PAX), RPMs and LF. The UCM model is a convenient way to additively decompose a time series into components: the trend, the seasons, the cycles, the autoregressive term, regressive terms involving lagged dependent variables, regressive terms on independent variable and the so-called irregular movements.The BSM is formally described by the equation

yt = μt + γt + εt where μt = μt-1 + βt-1 + ηt with βt = βt-1 + ξt

where ηt ~ niid(0,ση2) and ξt ~ niid (0,σξ

2).

The equation defining μt is called the level of the trend and the equation defining β t is called the (eventually stochastic) slope of the trend, the notation “niid” standing for normally independently and identically distributed. It is also assumed that ηt and ξt are independent of each other.

There are models for four separate entities: Domestic, Atlantic, Latin, and Pacific, corresponding to the U.S. Department of Transportation entity definitions used in Form 41 reporting. Overall a total of twelve sets of coefficients are developed, three sets of coefficients (one for the PAX model, one for the RPM model, and one for the LF model) for each of the four entities. Forecasts for ASMs and PTL for each entity are calculated using the forecasted values of RPMs and LF for ASMs and RPMs along with PAX for PTL. Forecasts for passenger yields are based on entity specific historic month over month variation applied to the latest actual monthly data for each entity as reported in the Airlines 4 America monthly yield report. For the remaining years, APO employs a three-stage, least squares (3SLS) regression analysis of a sys-tem of equations. The rationale behind choosing 3SLS over ordinary least squares (OLS) is that the er -rors of the different equations are correlated and 3SLS model provides a way to produce estimates that are more consistent and asymptotically efficient.6

For the 3SLS model, the following variables were used:

Endogenous variables7:

4 The modeling software used is SAS v9.3, copyrighted by SAS Institute Inc., Cary, NC USA.5 For further information, please see SAS/ETS 13.2 User’s Guide: The UCM Procedure, page 2304, Link to SAS.6 For more detailed information about the 3SLS model, please refer to SYSLIN procedure documentation at Link to SAS, SAS Institute Inc.7 Endogenous variables are variables determined by the system. Endogenous variables can also appear on the right-hand side of equations. Source: SAS Institute Inc. website, http://support.sas.com/documentation/cdl/en/etsug/60372/HTML/default/viewer.htm#etsug_syslin_sect004.htm, dated April 8, 2016.FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 6 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Log of mainline carrier RPMs Log of mainline carrier passenger yield Log of regional carrier load factor Log of mainline carrier load factor Log of mainline carrier real cost per available seat mile (ASM) Log of mainline carrier stage length

Instrumental variables8:

Log of personal consumption expenditure per capita Civilian unemployment rate Post September 11, 2001 dummy variable (fiscal year 2002 onwards) Mainline carrier’s share of domestic passenger market Regional carrier average passenger trip length Log of mainline carrier average passenger trip length A time variable (i.e., 1/(year – 1986)) Log of refiners acquisition cost (i.e., weighted average price of crude received in refinery)

The following relationships were then determined, and using the resultant coefficients, the dependent variables were forecast into the future.9 This procedure was done separately using mainline and regional carrier data to produce two sets of predicted variables.

8 Instrumental variables are predetermined variables used in obtaining values for the current period endogenous variables by a first-stage regression. The use of instrumental variables characterizes estimation methods such as two-stage least squares and three-stage least squares. Instrumental variables estimation methods substitutes these first-stage predicted values for endogenous variables when they appear as regressors in model equations. Source: SAS website, Link to SAS, dated 2014.9 The SIMLIN procedure was used to generate predicted values for the dependent variables using the coefficients that were produced by the SYSLIN procedure. For more detailed information, please refer to SIMLIN procedure documentation at http://support.sas.com/documentation/cdl/en/etsug/60372/HTML/default/viewer.htm#etsug_simlin_sect001.htm, SAS Institute Inc.FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 7 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Dependent variable Independent variables

Log of mainline carrier RPMs

Log of real PCE per capita

Unemployment rate

Log of mainline carrier passenger yield10

Post September 11, 2001 dummy variable

Log of mainline carrier real yield Log of mainline carrier passenger trip length

Log of mainline carrier real cost per ASM

Log of mainline carrier stage length Log of real refiners acquisition cost

Log of mainline carrier passenger trip length

Log of mainline carrier cost per ASM Log of mainline carrier stage length

Log of real refiners acquisition cost

Log of regional load factor

Time variable (i.e., 1/(year-1986))

Post September 11, 2001 dummy variable

Lagged log of regional load factor

Log of mainline carrier load factor

Time variable (i.e., 1/(year-1986))

Post September 11, 2001 dummy variable

Lagged log of mainline carrier load factor

These variables and the structure of the linear equations were chosen after much beta testing of different economic variables and model structures; this model produced the best fit and accurately reflected the analysts’ knowledge of the aviation industry. It will be subject to change in the future as the aviation in-dustry restructures itself or if major disruptions to the economy occur. The output from the statistical model is shown in Appendix C of this document.

For the Report, the growth rates of the statistical model’s predicted variables were used rather than the actual predicted values. The growth rates were spliced on to fiscal year 2016 estimates which were esti -mated separately via the BSM model described earlier. These forecast values were then used to generate the following forecast variables for mainline and re-gional carriers:

Forecast variable Formula11

Load factor RPMs / ASMs

Carrier departures Miles flown / stage length

Carrier miles flown Previous year value * growth rate of ASMs12

10 The term yield, in all of the domestic models detailed in this report, includes the following taxes and fees: ad valorem taxes, segment fee and TSA security fee.11 For ease of reading, multiplicative factors used to convert numbers to millions, thousands, etc. as needed have been eliminated.12 This growth rate was adjusted slightly by the analyst based on an understanding of industry trends.FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 8 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Carrier stage length Trip length / Trip vs stage length ratio

Seats per aircraft mile ASMs / miles flown

Mainline carrier passenger revenue Nominal passenger yield * RPMs

Mainline carrier nominal passenger yield Real passenger yield * consumer price index

Mainline carrier real passenger yieldPrevious year * statistical model’s predicted real

yield mainline carrier growth rate

Regional carrier passenger revenuePrevious year * (mainline real yield growth rate * re-

gional RPM growth rate)

Regional nominal passenger yield Passenger revenue / RPMs

Regional carrier real passenger yield Passenger revenue / consumer price index

Trip length versus stage length ratioAnnual growth rate of .05% was applied per analyst

judgment

The mainline and regional carrier variables are then summed to produce domestic totals; these numbers are reproduced in the various tables of Appendix C of the Report.

Alternative ScenariosOptimistic and pessimistic scenarios were also created for the domestic forecast. All of the model inputs, sources, and calculations are identical to the baseline forecast (described above) except for the economic data from IHS Global Insight.13 Rather, data from IHS Global Insight’s 10-year and 30-year optimistic and pessimistic forecasts from their January 2017 Baseline U.S. Economic Outlook were used. Inputs from these alternative scenarios were used to create a “high” and a “low” traffic, capacity, and yield forecast.

U.S. Airlines International Forecast

This forecast focuses solely on U.S. airlines flying into or out of the U.S. and relies upon Form 4114 data provided by BTS and IHS Global Insight. As is the case with the domestic forecast, it is a 20 year fore-cast based on the federal government’s fiscal year.

13 IHS Global Insight is a large, independent private consulting firm with a division devoted to economic analysis and forecasting. More information about the company can be found at Info from ihs.com.14 “The Form 41 Financial Reports contain financial information on large certificated U.S. air carriers. Financial information includes balance sheet, cash flow, employment, income statement, fuel cost and consumption, aircraft operating expenses, and operating expenses. This data is collected by the Office of Airline Information of the Bureau of Transportation Statistics. [Schedule P-1.2] provides quarterly profit and loss statements for carriers with annual operating revenues of $20 million or more. The data include operating revenues, operating expenses, depreciation and amortization, operating profit, income tax, and net income.” Data Profile: Air Carrier Statistics (Form 41 Traffic) for U.S. Carriers, BTS website, Info from BTS=.FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 9 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Forecast YearsThe Report includes historical data and forecast data for a 20 year horizon. Historical and forecast data presented include:

Economic assumptions Available seat miles (ASMs) Revenue passenger miles (RPMs) Load factor Nominal and real passenger yield15

Passengers Alternative (optimistic and pessimistic) scenarios

Data in the Report are presented on a U.S. Government fiscal year basis (October through September).Form 41 Forecast MethodologyHistorical data used to supply inputs into the forecast models were obtained from U.S. Department of Transportation’s Bureau of Transportation Statistics. Additional information about the input data can be found in Appendix B.

The statistical model16 used for the Form 41 based international forecast employs a general linear regres-sion model for three regions: Atlantic17, Latin18 and Pacific19. The dependent variable is RPMs for each model.

The independent variables for each model are shown below; additional information about them can be found in Appendix B.

15 For the international forecasts, real and nominal yield excludes taxes and fees due to airline reporting requirements on Form 41. 16 The modeling software used is SAS v9.3, copyrighted by SAS Institute Inc., Cary, NC USA.17 The Atlantic region includes Western and Central Europe, the Balkans, Commonwealth of Independent states, the Middle East, and Africa.18 The Latin region includes Latin America and the Caribbean.19 The Pacific region includes the Asia-Pacific region.FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 10 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Model Independent Variable Description

Atlantic region

US25For75 Ratio of indexed U.S. GDP to indexed Atlantic region GDP

Tension Gulf wars dummy variable; applied to 1991 and 2003

Post911Post September 11, 2001 dummy variable; applied to 2002-2036

Latin regionLatinGDPIx50 Ratio of indexed U.S. GDP to indexed Latin region GDP

Post911Post September 11, 2001 dummy variable; applied to 2002-2036

Pacific region

TotalPacAsiaGDPSum of U.S., Japan and Pacific region (excluding Japan) GDP

SARSSevere acute respiratory syndrome dummy variable; applied to 2003

GFC2Global financial crisis dummy variable; applied to 2008-2010

Post911Post September 11, 2001 dummy variable; applied to 2002-2036

These variables and the structure of the regression models were chosen after much testing of different economic variables and model structures; these models produced the best fit and accurately reflected the analysts’ knowledge of the aviation industry. They will be subject to change in the future as the aviation industry restructures itself or if major disruptions to the world economies occur. The output from the re-gional models is shown in Appendix C of this document.

The region specific models’ predicted annual growth rates for the dependent variable, RPMs, is then ap-plied to the last historical year of data; in this case, 2016. The final results are three forecasts of RPMs, one for each region.

To develop a forecast of passengers by region, the model’s forecast regional RPMs, described in the pre-ceding paragraph, are divided by an estimated annual trip length of the respective region. The latter is determined by an APO analyst looking at regional historical data and applying knowledge of the aviation industry. It should be noted that, globally, trip length is increasing at a decreasing rate since there is a natural limit to how far people are willingor needto fly on a single trip.

These forecast values were then used to generate the following forecast variables for mainline and re-gional carriers for each of the three regions:

Forecast variable Formula20

Nominal passenger revenue RPMs * Nominal yieldNominal yield Nominal passenger revenue / RPMsReal yield Nominal yield / CPI index

20 For ease of reading, multiplicative factors used to convert numbers to millions, thousands, etc. as needed have been eliminated.FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 11 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Seats per aircraft Forecast based on analyst judgment of historical trends and knowledge of the industry

Miles flown ASMs / Seats per aircraftTrip length RPMs / PassengersMainline trip vs stage length Forecast based on analyst judgment of historical

trends and knowledge of the industryMainline carrier stage length (miles) Total aircraft miles flown for all three regions /

Mainline trip vs stage length estimateMainline carrier departures Total miles flown for all three regions / Mainline

stage lengthRegional carrier international departures Forecast based on analyst judgment of historical

trends and knowledge of the industryTotal carrier departures Mainline + regional carrier departures for all three

regionsLoad factor RPMs / ASMs

Most of these variables are reproduced in the various tables of Appendix C of the Report.

Alternative ScenariosOptimistic and pessimistic scenarios were also created for the international F41 forecast. All of the model inputs, sources, and calculations are identical to the baseline forecast (described above) except for the economic data from IHS Global Insight. Rather, for U.S. GDP forecasts, data from IHS Global Insight’s 30-year optimistic and pessimistic forecasts from their September 2016 Baseline U.S. Economic Outlook were used. Since IHS Global Insight does not produce optimistic and pessimistic forecasts for their world GDP components table, a set of ratios were derived using Global Insight’s baseline, optimistic, and pes-simistic 30-year macro scenarios for Major Trading Partners GDP and Minor Trading Partners GDP. In-puts from these alternative scenarios were used to create a “high” and a “low” traffic, capacity, and yield forecast.

U.S. and Foreign Flag International Forecast

This passengers-only forecast includes U.S. and foreign flag carriers flying into or out of the U.S. and re -lies upon passenger data provided by the U.S. Customs and Border Protection (CBP) agency21 and GDP and exchange rate data provided by IHS Global Insight.

Forecast YearsThe Report includes historical data and forecast data for a 20 year horizon. Data in the Report are presented on a U.S. Government calendar year basis. CBP Forecast MethodologyHistorical data used to supply inputs into the forecast models were obtained from CBP. Additional infor-mation about the input data can be found in Appendix B.

21 Customs and border protection data is processed and released by the Department of Commerce.FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 12 of 63Methodology and Data Sources Office of Aviation Policy and Plans

The statistical model22 used for the CBP based international forecast employs a general linear regression model for multiple independent countries. These countries were chosen because they form the majority of the passengers traveling between the U.S. and foreign destinations. The dependent variable is pas-sengers for all of the models.

The independent variables for each model are shown below; additional information about them can be found in Appendix B. These models were chosen based on goodness of fit and the analyst’s knowledge of the aviation market within the country under review.

As is the case with the domestic forecast, this forecast is unconstrained as well.

22 The modeling software used is SAS v9.3, copyrighted by SAS Institute Inc., Cary, NC USA.FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 13 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Model Independent Variable DescriptionAtlantic Region

France

GDP5 Ratio of indexed U.S.GDP vs indexed France GDP

YieldForecast based on analyst judgment of historical trends and knowledge of the industry

Post911Post September 11, 2001 dummy variable; applied to 2002-2036

Germany

LGDP5 Log(ratio of indexed U.S. GDP vs indexed Germany GDP)

LExch Log(exchange rate of euro vs U.S. dollar)

Post911Post September 11, 2001 dummy variable; applied to 2002-2036

Ireland

LGDP6 Log(ratio of indexed U.S. GDP vs indexed Ireland GDP)

LExch Log(exchange rate of euro vs U.S. dollar)

YieldForecast based on analyst judgment of historical trends and knowledge of the industry

TravelTax Ireland Air Travel Tax dummy variable

Post911Post September 11, 2001 dummy variable; applied to 2002-2036

Italy

GDP7 Log(ratio of indexed U.S. GDP vs indexed Germany GDP)

PanAm Pan American bankruptcy dummy variable; applied to 1991

Post911Post September 11, 2001 dummy variable; applied to 2002-2036

IraqWar Iraq War dummy variable; applied to 2003

Millennium 2001 dummy variable

Netherlands

GDP5 Ratio of indexed U.S. GDP vs indexed Netherlands GDP

11-Sep September 11, 2001 dummy variable; applied to 2001

Post911Post September 11, 2001 dummy variable; applied to 2002-2036

Spain GDP1 Ratio of indexed U.S. GDP vs indexed Spain GDP

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 14 of 63Methodology and Data Sources Office of Aviation Policy and Plans

United Kingdom

GDP9 Ratio of indexed U.S. GDP vs indexed UK GDP

Post911Post September 11, 2001 dummy variable; applied to 2002-2036

GFCGlobal financial crisis dummy variable; applied to 2008-2036

Other European countries

GDP3Log(ratio of indexed U.S. GDP vs indexed other European countries GDP)

Post911Post September 11, 2001 dummy variable; applied to 2002-2036

11-Sep September 11, 2001 dummy variable; applied to 2001

Latin America Region

Bahamas

YieldForecast based on analyst judgment of historical trends and knowledge of the industry

Post911 Post September 11, 2001 dummy variable; applied to 2002-2036

11-Sep September 11, 2001 dummy variable; applied to 2001

Brazil

GDP4 Log(ratio of indexed U.S. GDP vs indexed Brazil GDP)11-Sep September 11, 2001 dummy variable; applied to 2001

Post911Post September 11, 2001 dummy variable; applied to 2002-2036

Dominican Republic

GDP5Log(ratio of indexed U.S. GDP vs indexed Dominican Republic GDP)

Jamaica GDP5 Log(ratio of indexed U.S. GDP vs indexed Jamaica GDP)

Mexico LGDP6 Log(ratio of indexed U.S. GDP vs indexed Mexico GDP)

Other Latin America countries

GDP3Log(ratio of indexed U.S. GDP vs indexed other Latin American countries GDP)

Post911Post September 11, 2001 dummy variable; applied to 2002-2036

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 15 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Pacific Region

China

GDP5 Ratio of indexed U.S. GDP vs indexed China GDPExch Exchange rate of Renminbi vs U.S. dollar

Post911Post September 11, 2001 dummy variable; applied to 2002-2036

Hong Kong

GDP3 Ratio of indexed U.S. GDP vs indexed Hong Kong GDPSARS SARS epidemic dummy variable; applied to 2003

Post911Post September 11, 2001 dummy variable; applied to 2002-2036

IndiaGDP5 Ratio of indexed U.S. GDP vs India indexed GDP

NonStopServStart of non-stop service from U.S. to India dummy variable; applied to 2006-2036

Japan

LGDP2 Log(ratio of indexed U.S. GDP vs indexed Japan GDP)

LNFlatYield Log of real yield held constant from 2015 onwards11-Sep September 11, 2001 dummy variable; applied to 2001

Post911Post September 11, 2001 dummy variable; applied to 2002-2036

South Korea

LGDP2Log(ratio of indexed U.S. GDP vs indexed South Korea GDP)

11-Sep September 11, 2001 dummy variable; applied to 2001

Post911Post September 11, 2001 dummy variable; applied to 2002-2036

FinanCrisis Financial crisis dummy variable; applied 1998-1999NWPaxData

Taiwan

GDP5 Ratio of indexed U.S. GDP vs indexed Taiwan GDP

Post911Post September 11, 2001 dummy variable; applied to 2002-2036

GFCGlobal financial crisis dummy variable; applied to 2008-2036

Other PacificGDP3

Ratio of indexed U.S. GDP vs indexed other Pacific countries GDP

GFCGlobal financial crisis dummy variable; applied to 2008-2036

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 16 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Transborder (via Canada)

LGDP7 Log(ratio of indexed U.S. GDP vs indexed Canada GDP)

11-Sep September 11, 2001 dummy variable; applied to 2001

Post911Post September 11, 2001 dummy variable; applied to 2002-2036

The passenger forecasts for the individual countries are not reported publicly; rather, only the annual totals for all countries combined are discussed in the text of the Report. The data are not represented in the tables in the appendices. Alternative forecasts for the CBP forecast are not done.

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 17 of 63Methodology and Data Sources Office of Aviation Policy and Plans

APPENDIX A: Glossary of terms

Acronym Description3SLS Three stage least square statistical modelAPO FAA Office of Aviation Policy and PlansASMs Available seat milesBSM Basic structural modelCBP U.S. Customs and Border Protection AgencyCY Calendar yearF41 or Form 41 Form 41 Financial Reports from the U.S. Bureau of Transportation StatisticsFAA Federal Aviation AdministrationFY Federal government fiscal year (October – September)GDP Gross domestic productOLS Ordinary least squares modelPAX PassengerPCE Personal consumption expenditurePTL Passenger trip lengthRAC Refiners acquisition costRPMs Revenue passenger milesSAS Statistical Analysis Software (a software suite developed by SAS Institute)SARS Severe acute respiratory syndrome

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 18 of 63Methodology and Data Sources Office of Aviation Policy and Plans

APPENDIX B: Data inputs and sources

Data inputs and sources for the baseline domestic forecast

Economic Variables (all data are converted to fiscal year by APO)Model Label Description Notes

Model input Source

CPIConsumer price index, all-urban, Source: BLS, Units: - 1982-84=1.00 seasonally adjusted

Index is used to calculate real prices, such as yield

Indirectly

IHS Global Insight, Mnemonic: Baseline: CPI.Q.FMS Optimistic: CPI.Q.FMBA2 Pessimistic: CPI.Q.FMBA1

UNRATECivilian unemployment rate Source: BLS Units: - percent

Yes

IHS Global Insight, Mnemonic: Baseline: RUC.Q.FMS Optimistic: RUC.Q.FMBA2 Pessimistic: RUC.Q.FMBA1

PCEReal Consumer Spending - Total personal consumption expenditures, Source: BEA, Units: Billion 2009 dollars annual rate. Variables are used to calculate

personal consumption expenditure per capita

Indirectly

IHS Global Insight, Mnemonic: Baseline: CONSR.Q.FMS Optimistic: CONSR.Q.FMBA2 Pessimistic: CONSR.Q.FMBA1

POPTotal population, including armed forces overseas Source: Census Units: millions- end of period

Indirectly

IHS Global Insight, Mnemonic: Baseline: NP.Q.FMS Optimistic: NP.Q.FMBA2 Pessimistic: NP.Q.FMBA1

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 19 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Model Label Description Notes

Model input Source

Log of PCEPC

Personal consumption expenditure per capita. APO-100 transforms data into natural log for model.

Is calculated by APO (PCE / Total population including armed forces overseas)

Yes APO

RAC

Refiners Acquisition Cost. Weighted average price of crude received in refinery inventories Source: DOE Units: dollars per barrel- not seasonally adjusted

Yes

IHS Global Insight, Mnemonic: Baseline: POILRAP.Q.FMS Optimistic: POILRAP.Q.FMBA2 Pessimistic.:

POILRAP.Q.FMBA1

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 20 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Aviation Variables (all data are converted to fiscal year by APO)

Model Label Description NotesModel input Source

Year Calendar year

Indirectly

Bureau of Transportation Statistics, TranStats, Form T1: U.S. Air Carrier Traffic And Capacity Summary by Service Class 23

Month Month of year

UniqueCarrierName

Unique Carrier Name. When the same name has been used by multiple carriers, a numeric suffix is used for earlier users, for example, Air Caribbean, Air Caribbean (1).

Each carrier is categorized as being either a network, regional, low fare or “other” carrier. All carriers are known “mainline” carriers with the exception of regionals.

UniqueCarrier

Unique Carrier Code. When the same code has been used by multiple carriers, a numeric suffix is used for earlier users, for example, PA, PA(1), PA(2). Use this field for analysis across a range of years.

CarrierRegion

Carrier's operation region. Carriers report data by operation region (Atlantic, Latin, Pacific, System, International, and Domestic)

For the domestic forecasts, Region = Domestic

T320_ASM Available seat miles Summed by airline category Yes

23 “The Air Carrier Statistics database, also known as the T-100 data bank, contains domestic and international airline market and segment data. Certificated U.S. air carriers report monthly air carrier traffic information using Form T-100. The data is collected by the Office of Airline Information, Bureau of Transportation Statistics, Research and Innovative Technology Administration. The tables in this database provide domestic market, domestic segment, international market, international segment, combined table for domestic and international market, combined table for domestic and international segment data by certificated U.S. air carriers. Large certificated carriers hold Certificates of Public Convenience and Necessity issued by the U.S. Department of Transportation authorizing the performance of air transportation with annual operating revenues of $20 million or more.” Data Profile: Air Carrier Statistics (Form 41 Traffic) for U.S. Carriers, BTS website, Info from BTS. FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 21 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Model Label Description NotesModel input Source

T140_RPM Revenue passenger milesT110_RPax Revenue passengers enplanedT410_RMilesFlown Revenue aircraft miles flown

Is used to calculate stage length IndirectlyT510_RDPerformed

Revenue aircraft departures performed

MainPTLComPTL

Mainline carrier passenger trip lengthRegional carrier passenger trip length

Historical data is calculated (RPMs/Passengers); future years is an exogenous variable decided by APO. These data are calculated separately for mainline and regional carriers.

Yes

MainPaxShrMainline carrier’s share of passenger market (versus the regional carriers)

Historical data is calculated; futures years is an exogenous variable decided by APO

Yes

MainStageAverage stage length for mainline carriers

Historical data is calculated (T410_RMilesFlown/ T510_RDPerformed) by APO

Yes

TotalEnplTotal passengers (mainline and regional)

Yes

MainLFComLF

Mainline and regional load factors Is calculated by APO (RPMs/ASMs) Yes

Log of MainYld2Average of mainline passenger yield transformed via natural log

Is calculated by APO [log(passenger revenue24 / mainline RPMs)]

Yes

MainCASM Average mainline cost per available seat mile

Is calculated by APO (mainline operating expenses / mainline ASMs)

Yes

24 Includes the following taxes: FAA ad valorem tax, segment fee and TSA security fee.FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 22 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Model Label Description NotesModel input Source

SvcClassType of service provided (scheduled, non-scheduled, etc.).

F designation was used for the domestic forecasts; that is, scheduled passenger/cargo service, can include freight or mail in the belly

Indirectly

LFOpexNetOpex

Low fare and Network carriers’ operating expenses

Is used to calculate mainline operating expenses

IndirectlyBureau of Transportation Statistics, TranStats, Form 41, Schedule P-1.2: Air Carrier Financial 25

LFPrevNetPrev

Low fare and Network carriers’ passenger revenue

Is used to calculate mainline passenger yield

Indirectly

Post911 Post 9/11 dummy variableApplied fiscal years 2002-2036 by APO

Yes

APOTime Time variable = 1/(year – 1986)

Used to dampen demand in the future as the aviation market reaches maturity

Yes

Data inputs and sources for the Form 41 baseline international forecastInfo from BTS

Economic Variables (all data are converted to fiscal year by APO)

25 “The Form 41 Financial Reports contain financial information on large certificated U.S. air carriers. Financial information includes balance sheet, cash flow, employment, income statement, fuel cost and consumption, aircraft operating expenses, and operating expenses. This data is collected by the Office of Airline Information of the Bureau of Transportation Statistics. [Schedule P-1.2] provides quarterly profit and loss statements for carriers with annual operating revenues of $20 million or more. The data include operating revenues, operating expenses, depreciation and amortization, operating profit, income tax, and net income.” Data Profile: Air Carrier Statistics (Form 41 Traffic) for U.S. Carriers, BTS website, From BTS=.

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 23 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Model label Description Notes

Model input Source

CPIConsumer price index, all-urban, Source: BLS, Units: - 1982-84=1.00 seasonally adjusted

Index is used to calculate real prices, such as yield

Indirectly

IHS Global Insight, Mnemonic: Baseline: CPI.Q.FMS Optimistic: CPI.Q.FMBA2 Pessimistic: CPI.Q.FMBA1

GDPReal annual GDP history and forecast estimates by country

A ratio of U.S. GDP to region specific GDP was developed for each of the region specific models (Atlantic, Latin and Pacific) by APO.

IndirectlyIHS Global Insight, GDP Components, Interim forecast, monthly, sheet GDPR$A

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 24 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Aviation Variables (all data are converted to fiscal year by APO)

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 25 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Model label Description NotesModel input Source

Passengers Mainline carrier passengers

International regional carrier passengers are grouped with the Latin region’s mainline carrier passengers

Indirectly Bureau of Transportation Statistics, TranStats, Form T1: U.S. Air Carrier Traffic And Capacity Summary by Service Class 26

Trip lengthAverage passenger trip length (PTL) in miles by region

Historical data is calculated via RPMs/Passengers by region. PTL is used to estimate regional passenger forecasts (RPMs / PTL) by APO.

Indirectly

Load Factor Average regional load factor

Historical data is calculated via RPMs / ASMs. Forecast load factor is estimated by the APO analyst based on knowledge of the aviation industry. Forecast load factor is used to forecast ASMs.

Indirectly APO

SARSSevere acute respiratory syndrome (SARS) dummy variable used in the Pacific region model

Applied fiscal year 2003 by APO Yes APO

26 “The Air Carrier Statistics database, also known as the T-100 data bank, contains domestic and international airline market and segment data. Certificated U.S. air carriers report monthly air carrier traffic information using Form T-100. The data is collected by the Office of Airline Information, Bureau of Transportation Statistics, Research and Innovative Technology Administration. The tables in this database provide domestic market, domestic segment, international market, international segment, combined table for domestic and international market, combined table for domestic and international segment data by certificated U.S. air carriers. Large certificated carriers hold Certificates of Public Convenience and Necessity issued by the U.S. Department of Transportation authorizing the performance of air transportation with annual operating revenues of $20 million or more.” Data Profile: Air Carrier Statistics (Form 41 Traffic) for U.S. Carriers, BTS website, From BTS. FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 26 of 63Methodology and Data Sources Office of Aviation Policy and Plans

APPENDIX C: Model outputs

Baseline Domestic Model Output

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 27 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Log (RPMs) = f(log PCEPPC, unemployment rate, log real yield, post911)

The SYSLIN ProcedureTwo-Stage Least Squares Estimation

Model MAINLINERPMDependent Variable lmainrpm

Label Log(Mainline RPMs)

Source DF Sum of Squares Mean Square F Value Pr > FModel 4 0.898262 0.224566 486 <.0001Error 24 0.01109 0.000462

Corrected Total 28 0.910331

Root MSE 0.0215 R-Square 0.9878Dependent Mean 12.98399 Adj R-Sq 0.98577

Coeff Var 0.16556

Parameter VariableEstimate Label

Intercept 1 -0.32901 1.183325 -0.28 0.7834 Interceptlpcepc 1 1.271901 0.132989 9.56 <.0001 Log(PCE per capita)

UNRATE 1 -0.01272 0.003505 -3.63 0.0013 Unemployment Ratelrmainyld2 1 -0.26701 0.135744 -1.97 0.0608 Log(Mainline Loaded Real Yield)POST911 1 -0.17338 0.023907 -7.25 <.0001 Post 9/11 dummy

Analysis of Variance

Parameter EstimatesVariable DF Standard Error t Value Pr > |t|

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 28 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Log (RPMs) = f(log PCEPPC, unemployment rate, log real yield, post911)

The SYSLIN ProcedureTwo-Stage Least Squares Estimation

Model MAINLINESTAGEDependent Variable lmainstage

Source DF Sum of Squares Mean Square F Value Pr > FModel 2 0.672492 0.336246 640.86 <.0001Error 26 0.013642 0.000525

Corrected Total 28 0.686133

Root MSE 0.02291 R-Square 0.98012Dependent Mean 6.65136 Adj R-Sq 0.97859

Coeff Var 0.34438

Parameter VariableEstimate Label

Intercept 1 -5.4748 0.485824 -11.27 <.0001 Interceptlrealrac 1 -0.05205 0.014491 -3.59 0.0013 log(Refiners Real Cost)lmainptl 1 1.808152 0.077776 23.25 <.0001 Log(Mailine Pax Trip Length)

Analysis of Variance

Parameter EstimatesVariable DF Standard Error t Value Pr > |t|

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 29 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Log (RPMs) = f(log PCEPPC, unemployment rate, log real yield, post911)

The SYSLIN ProcedureTwo-Stage Least Squares Estimation

Model MAINUNITCOSTDependent Variable lrmaincasm

Label Log(Mainline Real CASM)

Source DF Sum of Squares Mean Square F Value Pr > FModel 2 0.108143 0.054072 87.51 <.0001Error 26 0.016065 0.000618

Corrected Total 28 0.125803

Root MSE 0.02486 R-Square 0.87066Dependent Mean -1.87128 Adj R-Sq 0.86071

Coeff Var -1.32836

Parameter VariableEstimate Label

Intercept 1 0.240264 0.267549 0.9 0.3774 Interceptlmainstage 1 -0.42039 0.045781 -9.18 <.0001 Log(Mainline Stage)

lrealrac 1 0.179835 0.013645 13.18 <.0001 log(Refiners Real Cost)

Analysis of Variance

Parameter EstimatesVariable DF Standard Error t Value Pr > |t|

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 30 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Log (RPMs) = f(log PCEPPC, unemployment rate, log real yield, post911)

The SYSLIN ProcedureTwo-Stage Least Squares Estimation

Model COMMUTERLFDependent Variable lcomlf

Label Log(Commuter Load Factor)

Source DF Sum of Squares Mean Square F Value Pr > FModel 3 1.137124 0.379041 798.32 <.0001Error 25 0.01187 0.000475

Corrected Total 28 1.148994

Root MSE 0.02179 R-Square 0.98967Dependent Mean 4.12789 Adj R-Sq 0.98843

Coeff Var 0.52787

Parameter VariableEstimate Label

Intercept 1 0.628214 0.209987 2.99 0.0062 Intercepttime3 1 -0.11312 0.055518 -2.04 0.0523 Inverse of Time

POST911 1 0.045034 0.018577 2.42 0.0229 Post 9/11 dummylglcomlf 1 0.848856 0.052358 16.21 <.0001 Lag-log of Commuter Load

Factor

Analysis of Variance

Parameter EstimatesVariable DF Standard Error t Value Pr > |t|

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 31 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Log (RPMs) = f(log PCEPPC, unemployment rate, log real yield, post911)

The SYSLIN ProcedureTwo-Stage Least Squares Estimation

Model MAINLINELFDependent Variable lmainlf

Label Log(Mainline Load Factor)

Source DF Sum of Squares Mean Square F Value Pr > FModel 3 0.396461 0.132154 531.89 <.0001Error 25 0.006211 0.000248

Corrected Total 28 0.402672

Root MSE 0.01576 R-Square 0.98457Dependent Mean 4.28612 Adj R-Sq 0.98272

Coeff Var 0.36776

Parameter VariableEstimate Label

Intercept 1 0.598888 0.224742 2.66 0.0133 Intercepttime3 1 -0.09738 0.040144 -2.43 0.0228 Inverse of Time

POST911 1 0.020032 0.011003 1.82 0.0806 Post 9/11 dummylglmainlf 1 0.862436 0.05309 16.24 <.0001 Lag-log of Mainline Load Factor

Analysis of Variance

Parameter EstimatesVariable DF Standard Error t Value Pr > |t|

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 32 of 63Methodology and Data Sources Office of Aviation Policy and Plans

The SYSLIN ProcedureThree-Stage Least Squares Estimation

MAINLINERP MAINLINEYL MAINLINEST MAINUNITCO COMMUTERLF MAINLINELFMAINLINERP 0.000462 0.000255 0.000157 -0.000186 0.000132 0.000106MAINLINEYL 0.000255 0.002083 -0.000042 0.000126 0.000117 -0.000053MAINLINEST 0.000157 -0.000042 0.000525 -0.000162 0.000038 0.000047MAINUNITCO -0.000186 0.000126 -0.000162 0.000618 -0.00004 -0.000113

COMMUTERLF 0.000132 0.000117 0.000038 -0.00004 0.000475 0.000221MAINLINELF 0.000106 -0.000053 0.000047 -0.000113 0.000221 0.000248

MAINLINERP MAINLINEYL MAINLINEST MAINUNITCO COMMUTERLF MAINLINELFMAINLINERP 1 0.26027 0.3194 -0.34822 0.281 0.31412MAINLINEYL 0.26027 1 -0.0404 0.11074 0.11801 -0.07428MAINLINEST 0.3194 -0.0404 1 -0.28377 0.07546 0.13025MAINUNITCO -0.34822 0.11074 -0.28377 1 -0.07419 -0.288

COMMUTERLF 0.281 0.11801 0.07546 -0.07419 1 0.64423MAINLINELF 0.31412 -0.07428 0.13025 -0.288 0.64423 1

MAINLINERP MAINLINEYL MAINLINEST MAINUNITCO COMMUTERLF MAINLINELFMAINLINERP 1.47559 -0.44054 -0.33646 0.38574 -0.15174 -0.24356MAINLINEYL -0.44054 1.19135 0.11497 -0.17955 -0.24275 0.31658MAINLINEST -0.33646 0.11497 1.1685 0.20925 -0.01031 0.02894MAINUNITCO 0.38574 -0.17955 0.20925 1.30239 -0.24582 0.37169

COMMUTERLF -0.15174 -0.24275 -0.01031 -0.24582 1.84537 -1.22867MAINLINELF -0.24356 0.31658 0.02894 0.37169 -1.22867 1.99484

MAINLINERP MAINLINEYL MAINLINEST MAINUNITCO COMMUTERLF MAINLINELFMAINLINERP 3193.42 -449.091 -683.34 721.92 -323.96 -718.82MAINLINEYL -449.09 572.067 109.98 -158.28 -244.12 440.11MAINLINEST -683.34 109.985 2227.09 367.51 -20.66 80.14MAINUNITCO 721.92 -158.284 367.51 2107.83 -453.85 948.64

COMMUTERLF -323.96 -244.121 -20.66 -453.85 3886.62 -3577.25MAINLINELF -718.82 440.109 80.14 948.64 -3577.25 8028.84

System Weighted MSE 1.0257Degrees of freedom 152

System Weighted R-Square 0.9754

Cross Model Covariance

Cross Model Correlation

Cross Model Inverse Correlation

Cross Model Inverse Covariance

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 33 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Model MAINLINERPDependent Variable lmainrpm

Label Log(Mainline RPMs)

Parameter VariableEstimate Label

Intercept 1 0.466364 1.006635 0.46 0.6473 Interceptlpcepc 1 1.19298 0.112947 10.56 <.0001 Log(PCE per capita)

UNRATE 1 -0.01149 0.002999 -3.83 0.0008 Unemployment Ratelrmainyld2 1 -0.26285 0.115661 -2.27 0.0323 Log(Mainline Loaded Real Yield)POST911 1 -0.14762 0.020635 -7.15 <.0001 Post 9/11 dummy

Durbin-Watson 0.851783Number of Observations 29

First-Order Autocorrelation 0.536226

Parameter EstimatesVariable DF Standard Error t Value Pr > |t|

Model MAINLINEYLDependent Variable lrmainyld2

Label Log(Mainline Loaded Real Yield)

Parameter VariableEstimate Label

Intercept 1 12.98671 0.71077 18.27 <.0001 Interceptlmainptl 1 -1.99649 0.090156 -22.14 <.0001 Log(Mailine Pax Trip Length)

lrmaincasm 1 0.539569 0.134514 4.01 0.0005 Log(Mainline Real CASM)

Durbin-Watson 0.693547Number of Observations 29

First-Order Autocorrelation 0.564376

Parameter EstimatesVariable DF Standard Error t Value Pr > |t|

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 34 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Model MAINLINESTDependent Variable lmainstage

Parameter VariableEstimate Label

Intercept 1 -5.3349 0.474337 -11.25 <.0001 Interceptlrealrac 1 -0.04705 0.013884 -3.39 0.0022 log(Refiners Real Cost)lmainptl 1 1.784831 0.075735 23.57 <.0001 Log(Mailine Pax Trip Length)

Durbin-Watson 0.607206Number of Observations 29

First-Order Autocorrelation 0.615985

Parameter EstimatesVariable DF Standard Error t Value Pr > |t|

Model MAINUNITCODependent Variable lrmaincasm

Label Log(Mainline Real CASM)

Parameter VariableEstimate Label

Intercept 1 0.135043 0.258151 0.52 0.6053 Interceptlmainstage 1 -0.40202 0.043744 -9.19 <.0001 Log(Mainline Stage)

lrealrac 1 0.175383 0.012621 13.9 <.0001 log(Refiners Real Cost)

Durbin-Watson 1.693041Number of Observations 29

First-Order Autocorrelation 0.152279

Parameter EstimatesVariable DF Standard Error t Value Pr > |t|

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 35 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Model COMMUTERLFDependent Variable lcomlf

Label Log(Commuter Load Factor)

Parameter VariableEstimate Label

Intercept 1 0.713747 0.199917 3.57 0.0015 Intercepttime3 1 -0.09851 0.053694 -1.83 0.0785 Inverse of Time

POST911 1 0.055976 0.017945 3.12 0.0045 Post 9/11 dummylglcomlf 1 0.826296 0.049872 16.57 <.0001 Lag-log of Commuter Load

Factor

Durbin-Watson 1.786432Number of Observations 29

First-Order Autocorrelation 0.100357

Parameter EstimatesVariable DF Standard Error t Value Pr > |t|

Model MAINLINELFDependent Variable lmainlf

Label Log(Mainline Load Factor)

Parameter VariableEstimate Label

Intercept 1 0.615186 0.207218 2.97 0.0065 Intercepttime3 1 -0.07206 0.03771 -1.91 0.0675 Inverse of Time

POST911 1 0.024222 0.010352 2.34 0.0276 Post 9/11 dummylglmainlf 1 0.857506 0.048962 17.51 <.0001 Lag-log of Mainline Load Factor

Durbin-Watson 2.314348Number of Observations 29

First-Order Autocorrelation -0.1601

Parameter EstimatesVariable DF Standard Error t Value Pr > |t|

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 36 of 63Methodology and Data Sources Office of Aviation Policy and Plans

The SIMLIN Procedure

Variable lmainrpm lrmainyld2 lmainstage lrmaincasm lcomlf lmainlflmainrpm 1 -0.2628 0.057 -0.1418 0 0lrmainyld2 0 1 -0.2169 0.5396 0 0

lcomlf 0 0 0 0 1 0lmainlf 0 0 0 0 0 1

lrmaincasm 0 0 -0.402 1 0 0lmainstage 0 0 1 0 0 0

Variable lglmainlf lglcomlflmainrpm 0 0lrmainyld2 0 0

lcomlf 0 0.8263lmainlf 0.8575 0

lrmaincasm 0 0lmainstage 0 0

Variable lpcepc UNRATE POST911 lmainptl lrealrac time3 MAINPAXSHR

Intercept

lmainrpm 1.193 -0.0115 -0.1476 0.6265 -0.0276 0 0 -3.2705lrmainyld2 0 0 0 -2.3837 0.1048 0 0 14.2168

lcomlf 0 0 0.056 0 0 -0.0985 0 0.7137lmainlf 0 0 0.0242 0 0 -0.0721 0 0.6152

lrmaincasm 0 0 0 -0.7175 0.1943 0 0 2.2798lmainstage 0 0 0 1.7848 -0.047 0 0 -5.3349

Inverse Coefficient Matrix for Endogenous Variables

Reduced Form for LaggedEndogenous Variables

Reduced Form for Exogenous Variables

The SIMLIN Procedure

Mean Pct Mean Abs RMS RMS PctError Pct Error Error Error

lmainrpm 29 -7.23E-15 -0.00058 0.0158 0.12253 0.0221 0.1714 Log(Mainline RPMs)lrmainyld2 29 8.96E-16 -0.0829 0.0379 2.31347 0.0468 2.7903 Log(Mainline Loaded Real Yield)

lcomlf 29 0.000392 -0.00667 0.0269 0.66329 0.0333 0.8278 Log(Commuter Load Factor)lmainlf 29 -0.000579 -0.0187 0.0141 0.33319 0.0182 0.4331 Log(Mainline Load Factor)

lrmaincasm 29 -8.81E-16 -0.024 0.0215 1.14361 0.0276 1.4582 Log(Mainline Real CASM)lmainstage 29 -3.28E-15 -0.001238 0.0185 0.28057 0.0217 0.3319 Log(Mainline Stage)

Fit StatisticsVariable N Mean Error Mean Abs Error Label

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 37 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Baseline International (Form 41) Model Output

Pacific RegionThe REG ProcedureModel: MODEL1Dependent Variable: RPMs RPMs

Number of Observations Read

41

Number of Observations Used

17

Number of Observations with

Missing Values

24

Sum of MeanSquares Square

Model 4 814767089 203691772 52.8 <.0001Error 12 46292321 3857693

Corrected Total 16 861059410

Root MSE 1964.10117 R-Square 0.9462Dependent Mean 60779 Adj R-Sq 0.9283

Coeff Var 3.23153

Parameter StandardEstimate Error

Intercept Intercept 1 25537 3216.633 7.94 <.0001TotalPacAsiaGDP TotalPacAsi

aGDP1 1.33481 0.11605 11.5 <.0001

SARS SARS 1 -7205.4685 2220.469 -3.25 0.007GFC2 GFC2 1 -3363.75852 1287.947 -2.61 0.0227

Post911 Post911 1 -7977.72872 1847.429 -4.32 0.001

The REG ProcedureModel: MODEL1Dependent Variable: RPMs RPMs

Durbin-Watson D 1.859Number of

Observations17

1st Order Autocorrelation

0.067

Parameter EstimatesVariable Label DF t Value Pr > |t|

Analysis of VarianceSource DF F Value Pr > F

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 38 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Atlantic RegionThe AUTOREG Procedure

Dependent Variable RPMsRPMs

SSE 691974007 DFE 23MSE 30085826 Root MSE 5485SBC 550.405025 AIC 545.221677MAE 3695.73353 AICC 547.039859

MAPE 4.04976102 HQC 546.762958Total R-Square 0.9373

Order DW1 0.56932 1.134

Standard ApproxError Pr > |t|

Intercept 1 -36580 12117 -3.02 0.0061US25For75 1 1446 163.5678 8.84 <.0001 US25For75

Tension 1 -10267 4372 -2.35 0.0278 TensionPost911 1 -6596 4814 -1.37 0.1839 Post911

Lag Covariance Correlation -1 9 8 7 6 5 4 3 2 1 0 1 2 3 4 5 6 7 8 9 

1 0 25628667 1 |                    |********************|1 15983282 0.623649 |                    |************        |

Preliminary MSE 15660715

StandardError

1 -0.623649 0.16666 -3.74

SSE 321976919 DFE 22MSE 14635315 Root MSE 3826SBC 533.536567 AIC 527.057383MAE 2619.12544 AICC 529.914526

MAPE 2.99926709 HQC 528.983984Transformed

Regression R-Square

0.8658

Total R-Square 0.9708

Order DW1 0.96132 1.4346

Standard ApproxError Pr > |t|

Intercept 1 -40236 12250 -3.28 0.0034US25For75 1 1506 155.8488 9.66 <.0001 US25For75

Tension 1 -6940 2310 -3 0.0065 TensionPost911 1 -10995 4147 -2.65 0.0146 Post911

Yule-Walker Estimates

Durbin-Watson Statistics

Parameter EstimatesVariable DF Estimate t Value Variable Label

Estimates of Autocorrelations

Estimates of Autoregressive ParametersLag Coefficient t Value

Ordinary Least Squares Estimates

Durbin-Watson Statistics

Parameter EstimatesVariable DF Estimate t Value Variable Label

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 39 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Latin RegionThe AUTOREG Procedure

Dependent Variable RPMsRPMs

SSE 45228065.5 DFE 14MSE 3230576 Root MSE 1797SBC 308.241806 AIC 305.742166MAE 1151.08522 AICC 307.588319

MAPE 1.879523 HQC 305.990635Total R-Square 0.9902

Order DW1 0.77962 1.4523

Standard ApproxError Pr > |t|

Intercept 1 -88948 3965 -22.43 <.0001LatinGDPIx50 1 1577 47.3669 33.28 <.0001 LatinGDPIx50

Post911 1 -6188 1620 -3.82 0.0019 Post911

Lag Covariance Correlation -1 9 8 7 6 5 4 3 2 1 0 1 

2 3 4 5 6 7 8 9 1 0 2660474 1 |                    |**********

**********|1 979819 0.368287 |                    |*******     

        |

Preliminary MSE 2299619

StandardError

1 -0.368287 0.257856 -1.43

SSE 35507427.3 DFE 13MSE 2731341 Root MSE 1653SBC 307.107189 AIC 303.774336MAE 1118.23468 AICC 307.107669

MAPE 1.88998277 HQC 304.105628Transformed

Regression R-Square

0.984

Total R-Square 0.9923

Order DW1 0.95742 1.4177

Standard ApproxError Pr > |t|

Intercept 1 -89637 5289 -16.95 <.0001LatinGDPIx50 1 1580 60.7988 26 <.0001 LatinGDPIx50

Post911 1 -5649 1704 -3.32 0.0056 Post911

Durbin-Watson Statistics

Parameter EstimatesVariable DF Estimate t Value Variable Label

Ordinary Least Squares Estimates

Durbin-Watson Statistics

Parameter EstimatesVariable DF Estimate t Value Variable Label

Yule-Walker Estimates

Estimates of Autoregressive Parameters

Estimates of Autocorrelations

Lag Coefficient t Value

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 40 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Baseline International (Customs and Border Protection) Model OutputFranceThe REG ProcedureModel: MODEL1Dependent Variable: PaxFrance PaxFrance

Number of Observations Read

53

Number of Observations Used

27

Number of Observations with

Missing Values

26

Sum of MeanSquares Square

Model 3 3.51E+13 1.17E+13 340.98 <.0001Error 23 7.88E+11 34268801046

Corrected Total 26 3.58E+13

Root MSE 185118 R-Square 0.978Dependent Mean 5370533 Adj R-Sq 0.9751

Coeff Var 3.44693

Parameter Standard

Estimate ErrorIntercept Intercept 1 -1377779 677054 -2.03 0.0535

gdp5 1 100625 5717.34 17.6 <.0001Post911 Post911 1 -857546 153618 -5.58 <.0001

Yield Yield 1 -124471 28268 -4.4 0.0002

Analysis of VarianceSource DF F Value Pr > F

Parameter EstimatesVariable Label DF t Value Pr > |t|

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 41 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Germany The REG ProcedureModel: MODEL1Dependent Variable: lnPaxGermany

Number of Observations Read

51

Number of Observations Used

27

Number of Observations with

Missing Values

24

Sum of MeanSquares Square

Model 3 1.6572 0.5524 404.49 <.0001Error 23 0.03141 0.00137

Corrected Total 26 1.68861

Root MSE 0.03695 R-Square 0.9814Dependent Mean 15.8437 Adj R-Sq 0.979

Coeff Var 0.23325

Parameter StandardEstimate Error

Intercept Intercept 1 6.8745 0.43822 15.69 <.0001lgdp5 1 1.99658 0.10002 19.96 <.0001lexch 1 -0.39131 0.0645 -6.07 <.0001

Post911 Post911 1 -0.1382 0.03144 -4.4 0.0002

Analysis of VarianceSource DF F Value Pr > F

Parameter EstimatesVariable Label DF t Value Pr > |t|

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 42 of 63Methodology and Data Sources Office of Aviation Policy and Plans

IrelandThe REG ProcedureModel: MODEL1Dependent Variable: lnPaxIreland

Number of Observations Read

51

Number of Observations Used

27

Number of Observations with

Missing Values

24

Sum of MeanSquares Square

Model 5 3.92622 0.78524 82.54 <.0001Error 21 0.19979 0.00951

Corrected Total 26 4.12601

Root MSE 0.09754 R-Square 0.9516Dependent Mean 14.28816 Adj R-Sq 0.94

Coeff Var 0.68265

Parameter StandardEstimate Error

Intercept Intercept 1 11.07946 0.57301 19.34 <.0001lgdp6 1 0.93331 0.10451 8.93 <.0001lexch 1 -0.70205 0.23612 -2.97 0.0073

Post911 Post911 1 -0.23031 0.09919 -2.32 0.0304Yield Yield 1 -0.05789 0.01834 -3.16 0.0048

TravelTax TravelTax 1 -0.17231 0.05736 -3 0.0068

Analysis of VarianceSource DF F Value Pr > F

Parameter EstimatesVariable Label DF t Value Pr > |t|

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 43 of 63Methodology and Data Sources Office of Aviation Policy and Plans

ItalyThe AUTOREG Procedure

Dependent Variable PaxItalyPaxItaly

SSE 3.92E+11 DFE 21MSE 1.87E+10 Root MSE 136694SBC 728.189148 AIC 720.414127MAE 100492.301 AICC 724.614127

MAPE 4.13081381 HQC 722.726048Total R-Square 0.9441

Order DW1 0.71682 1.4168

Standard ApproxError Pr > |t|

Intercept 1 -1733265 386377 -4.49 0.0002gdp7 1 48610 5006 9.71 <.0001

PanAm 1 -135220 150625 -0.9 0.3795 PanAmMillenium 1 310553 155302 2 0.0586 MilleniumPost911 1 -408483 133307 -3.06 0.0059 Post911IraqWar 1 -413986 147339 -2.81 0.0105 IraqWar

Lag Covariance Correlation -1 9 8 7 6 5 4 3 2 1 0 1 2 3 4 5 6 7 8 9 1

 0 1.45E+10 1 |                    |********************|1 8.27E+09 0.568847 |                    |***********         |

Preliminary MSE 9.83E+09

StandardError

1 -0.568847 0.183904 -3.09

SSE 1.92E+11 DFE 20MSE 9579857985 Root MSE 97877SBC 712.520738 AIC 703.44988MAE 68436.9018 AICC 709.344617

MAPE 2.81080663 HQC 706.147122Transformed

Regression R-Square

0.9231

Total R-Square 0.9727

Order DW1 0.87772 1.6758

Standard ApproxError Pr > |t|

Intercept 1 -1877248 353665 -5.31 <.0001gdp7 1 50474 4370 11.55 <.0001

PanAm 1 -178715 86106 -2.08 0.0511 PanAmMillenium 1 436072 86030 5.07 <.0001 MilleniumPost911 1 -447007 105728 -4.23 0.0004 Post911IraqWar 1 -266233 85546 -3.11 0.0055 IraqWar

Yule-Walker Estimates

Durbin-Watson Statistics

Parameter EstimatesVariable DF Estimate t Value Variable Label

Ordinary Least Squares Estimates

Durbin-Watson Statistics

Parameter EstimatesVariable DF Estimate t Value Variable Label

Estimates of Autocorrelations

Estimates of Autoregressive ParametersLag Coefficient t Value

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 44 of 63Methodology and Data Sources Office of Aviation Policy and Plans

NetherlandsThe AUTOREG Procedure

Dependent Variable PaxNetherlandsPaxNetherlands

SSE 2.47E+12 DFE 23MSE 1.07E+11 Root MSE 327862SBC 771.295561 AIC 766.112214MAE 252788.026 AICC 767.930395

MAPE 7.85122523 HQC 767.653494Total R-Square 0.895

Order DW1 0.6732 1.2868

Standard ApproxError Pr > |t|

Intercept 1 -3836323 689801 -5.56 <.0001gdp5 1 96451 9505 10.15 <.0001

11-Sep 1 -643800 374515 -1.72 0.099 11-SepPost911 1 -1418472 293755 -4.83 <.0001 Post911

Lag Covariance Correlation -1 9 8 7 6 5 4 3 2 1 0 1 2 3 4 5 6 7 8 9 1

 0 9.16E+10 1 |                    |********************|1 5.38E+10 0.587589 |                    |************        |

Preliminary MSE 6.00E+10

StandardError

1 -0.587589 0.172513 -3.41

SSE 1.36E+12 DFE 22MSE 6.17E+10 Root MSE 248469SBC 758.841966 AIC 752.362782MAE 178265.114 AICC 755.219925

MAPE 5.33672468 HQC 754.289383Transformed

Regression R-Square

0.8078

Total R-Square 0.9423

Order DW1 1.38092 1.949

Standard ApproxError Pr > |t|

Intercept 1 -3444748 835901 -4.12 0.0004gdp5 1 89703 11113 8.07 <.0001

11-Sep 1 -726573 276272 -2.63 0.0153 11-SepPost911 1 -1122469 337467 -3.33 0.0031 Post911

Yule-Walker Estimates

Durbin-Watson Statistics

Parameter EstimatesVariable DF Estimate t Value Variable Label

Estimates of Autocorrelations

Estimates of Autoregressive ParametersLag

Ordinary Least Squares Estimates

Durbin-Watson Statistics

Parameter EstimatesVariable DF Estimate t Value Variable Label

Coefficient t Value

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 45 of 63Methodology and Data Sources Office of Aviation Policy and Plans

SpainThe AUTOREG Procedure

Dependent Variable PaxSpainPaxSpain

SSE 4.72E+12 DFE 25MSE 1.89E+11 Root MSE 434664SBC 782.1821 AIC 779.590465MAE 334993.7 AICC 780.090465

MAPE 16.87461 HQC 780.361106Total R-Square

0.666

Order DW1 0.16432 0.4409

Standard ApproxError Pr > |t|

Intercept 1 -1441317 477561 -3.02 0.0058gdp1 1 39136 5543 7.06 <.0001

Lag Covariance

Correlation -1 9 8 7 6 5 4 3 2 1 0 1 2 3 4 5 6 7 8 9 1 

0 1.75E+11 1 |                    |********************|

1 1.48E+11 0.84777 |                    |*****************   |

Preliminary MSE 4.92E+10

StandardError

1 -0.84777 0.10826 -7.83

SSE 8.36E+11 DFE 24MSE 3.48E+10 Root MSE 186619SBC 739.987 AIC 736.099501MAE 138447 AICC 737.14298

MAPE 7.583773 HQC 737.255462Transformed Regression R-Square

0.4207

Total R-Square

0.9409

Order DW1 0.99132 1.3006

Standard ApproxError Pr > |t|

Intercept 1 -1296932 810993 -1.6 0.1229gdp1 1 39070 9359 4.17 0.0003

Yule-Walker Estimates

Durbin-Watson Statistics

Parameter EstimatesVariable DF Estimate t Value

Estimates of Autocorrelations

Estimates of Autoregressive ParametersLag Coefficie

ntt Value

Ordinary Least Squares Estimates

Durbin-Watson Statistics

Parameter EstimatesVariable DF Estimate t Value

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 46 of 63Methodology and Data Sources Office of Aviation Policy and Plans

United KingdomThe REG ProcedureModel: MODEL1Dependent Variable: PaxUK PaxUK

Number of Observations Read 51Number of Observations Used 27Number of Observations with

Missing Values24

Sum of MeanSquares Square

Model 3 1.73E+14 5.77E+13 71.25 <.0001Error 23 1.86E+13 8.10E+11

Corrected Total 26 1.92E+14

Root MSE 900198 R-Square 0.9028Dependent Mean 15634385 Adj R-Sq 0.8902

Coeff Var 5.75781

Parameter StandardEstimate Error

Intercept Intercept 1 -6756266 1857003 -3.64 0.0014gdp9 1 285735 25522 11.2 <.0001

Post911 Post911 1 -3539837 730143 -4.85 <.0001GFC GFC 1 -2252933 538051 -4.19 0.0004

Label DF t Value Pr > |t|

Analysis of VarianceSource DF F Value Pr > F

Parameter EstimatesVariable

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 47 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Other European CountriesThe AUTOREG Procedure

Dependent Variable PaxOtherEuropePaxOtherEurope

SSE 4.72E+13 DFE 18MSE 2.62E+12 Root MSE 1618840SBC 699.460396 AIC 695.096227MAE 1088471.52 AICC 697.449168

MAPE 10.6053476 HQC 696.124294Total R-Square 0.8939

Order DW1 0.5552 1.4459

Standard ApproxError Pr > |t|

Intercept 1 -20335649 2725702 -7.46 <.0001gdp3 1 424924 38967 10.9 <.0001

Post911 1 -8485590 1402495 -6.05 <.0001 Post91111-Sep 1 -3558566 1775471 -2 0.0603 11-Sep

Lag Covariance Correlation -1 9 8 7 6 5 4 3 2 1 0 1 2 3 4 5 

6 7 8 9 1 0 2.14E+12 1 |                    |**

******************|

1 1.29E+12 0.599347 |                    |************        |

Preliminary MSE 1.37E+12

StandardError

1 -0.599347 0.194147 -3.09

SSE 2.29E+13 DFE 17MSE 1.34E+12 Root MSE 1159430SBC 687.052632 AIC 681.59742MAE 805099.94 AICC 685.34742

MAPE 7.69865646 HQC 682.882504Transformed

Regression R-Square

0.8147

Total R-Square 0.9486

Order DW1 0.6282 1.3019

Standard ApproxError Pr > |t|

Intercept 1 -17576840 3474779 -5.06 <.0001gdp3 1 379157 46718 8.12 <.0001

Post911 1 -6207535 1593633 -3.9 0.0012 Post91111-Sep 1 -2847873 1259370 -2.26 0.0371 11-Sep

Estimates of Autocorrelations

Estimates of Autoregressive Parameters

Parameter Estimates

Ordinary Least Squares Estimates

Durbin-Watson Statistics

Parameter EstimatesVariable DF Estimate t Value Variable Label

Lag Coefficient t Value

Yule-Walker Estimates

Durbin-Watson Statistics

Variable DF Estimate t Value Variable Label

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 48 of 63Methodology and Data Sources Office of Aviation Policy and Plans

BahamasThe AUTOREG Procedure

Dependent Variable PaxBahamasPaxBahamas

SSE 5.49E+11 DFE 23MSE 2.39E+10 Root MSE 154523SBC 730.674007 AIC 725.490659MAE 104874.811 AICC 727.308841

MAPE 4.09252408 HQC 727.03194Total R-Square 0.2818

Order DW1 0.92472 1.4538

Standard ApproxError Pr > |t|

Intercept 1 3352148 358786 9.34 <.0001YldRlBaha2 1 -31860 13267 -2.4 0.0248 YldRlBaha2

Post911 1 -167572 128862 -1.3 0.2063 Post91111-Sep 1 -177214 172267 -1.03 0.3143 11-Sep

Lag Covariance Correlation -1 9 8 7 6 5 4 3 2 1 0 1 2 3 4 5 6 7 8 9 1 

0 2.03E+10 1 |                    |********************|

1 7.91E+09 0.388809 |                    |********            |

Preliminary MSE 1.73E+10

StandardError

1 -0.388809 0.196426 -1.98

SSE 4.27E+11 DFE 22MSE 1.94E+10 Root MSE 139267SBC 727.320342 AIC 720.841158MAE 96299.331 AICC 723.698301

MAPE 3.77747988 HQC 722.767759Transformed

Regression R-Square

0.2971

Total R-Square 0.4419

Order DW1 1.50012 1.4382

Standard ApproxError Pr > |t|

Intercept 1 3629999 380595 9.54 <.0001YldRlBaha2 1 -41570 14228 -2.92 0.0079 YldRlBaha2

Post911 1 -264655 141706 -1.87 0.0752 Post91111-Sep 1 -172825 148108 -1.17 0.2558 11-Sep

Parameter EstimatesVariable DF Estimate t Value Variable Label

Lag Coefficient t Value

Yule-Walker Estimates

Durbin-Watson Statistics

Estimates of Autoregressive Parameters

Estimates of Autocorrelations

Ordinary Least Squares Estimates

Durbin-Watson Statistics

Parameter EstimatesVariable DF Estimate t Value Variable Label

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 49 of 63Methodology and Data Sources Office of Aviation Policy and Plans

BrazilThe AUTOREG Procedure

Dependent Variable PaxBrazilPaxBrazil

SSE 3.03E+12 DFE 18MSE 1.69E+11 Root MSE 410589SBC 639.098054 AIC 634.733884MAE 310870.172 AICC 637.086825

MAPE 11.2943113 HQC 635.761951Total R-Square 0.8841

Order DW1 0.68632 1.4986

Standard ApproxError Pr > |t|

Intercept 1 -5196987 736289 -7.06 <.0001gdp4 1 108963 10224 10.66 <.0001

11-Sep 1 -984073 448551 -2.19 0.0416 11-SepPost911 1 -1965223 331813 -5.92 <.0001 Post911

Lag Covariance Correlation -1 9 8 7 6 5 4 3 2 1 0 1 2

 3 4 5 6 7 8 9 1 0 1.38E+11 1 |                    |***********

*********|1 8.86E+10 0.642123 |                    |***********

**       |

Preliminary MSE 8.11E+10

StandardError

1 -0.642123 0.185928 -3.45

SSE 1.48E+12 DFE 17MSE 8.72E+10 Root MSE 295367SBC 626.970739 AIC 621.515527MAE 221622.57 AICC 625.265527

MAPE 7.81293265 HQC 622.800611Transformed

Regression R-Square

0.7367

Total R-Square 0.9433

Order DW1 1.04762 1.512

Standard ApproxError Pr > |t|

Intercept 1 -4003410 1043696 -3.84 0.0013gdp4 1 90330 13653 6.62 <.0001

11-Sep 1 -624233 316627 -1.97 0.0652 11-SepPost911 1 -1275257 404579 -3.15 0.0058 Post911

Estimates of Autoregressive ParametersLag Coefficient t Value

Yule-Walker Estimates

Durbin-Watson Statistics

Parameter EstimatesVariable DF Estimate t Value Variable Label

Estimates of Autocorrelations

Ordinary Least Squares Estimates

Durbin-Watson Statistics

Parameter EstimatesVariable DF Estimate t Value Variable Label

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 50 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Dominican RepublicThe AUTOREG Procedure

Dependent Variable lnPaxDomRep

SSE 0.24553603 DFE 25MSE 0.00982 Root MSE 0.0991SBC -43.689654 AIC -46.281328MAE 0.08103574 AICC -45.781328

MAPE 0.53914218 HQC -45.510687Total R-Square 0.9464

Order DW1 0.32872 0.7116

Standard ApproxError Pr > |t|

Intercept 1 8.903 0.2891 30.79 <.0001lgdp5 1 1.3916 0.0663 21 <.0001

Lag Covariance Correlation -1 9 8 7 6 5 4 3 2 1 0 1 2 3 4 5 

6 7 8 9 1 0 0.00909 1 |                    |**

******************|

1 0.00689 0.75757 |                    |***************     |

Preliminary MSE 0.00387

StandardError

1 -0.75757 0.133243 -5.69

SSE 0.08118447 DFE 24MSE 0.00338 Root MSE 0.05816SBC -69.422138 AIC -73.309648MAE 0.04482393 AICC -72.26617

MAPE 0.30026649 HQC -72.153688Transformed

Regression R-Square

0.852

Total R-Square 0.9823

Order DW1 1.49382 1.677

Standard ApproxError Pr > |t|

Intercept 1 9.0257 0.5085 17.75 <.0001lgdp5 1 1.3693 0.1165 11.76 <.0001

Yule-Walker Estimates

Durbin-Watson Statistics

Parameter EstimatesVariable DF Estimate t Value

Estimates of Autocorrelations

Estimates of Autoregressive ParametersLag Coefficient t Value

Ordinary Least Squares Estimates

Durbin-Watson Statistics

Parameter EstimatesVariable DF Estimate t Value

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 51 of 63Methodology and Data Sources Office of Aviation Policy and Plans

JamaicaThe AUTOREG Procedure

Dependent Variable lnPaxJamaica

SSE 0.14926337 DFE 25MSE 0.00597 Root MSE 0.07727SBC -57.128401 AIC -59.720075MAE 0.06068505 AICC -59.220075

MAPE 0.4113242 HQC -58.949434Total R-Square

0.8755

Order DW1 0.47082 0.7918

Standard ApproxError Pr > |t|

Intercept 1 7.4295 0.5498 13.51 <.0001lgdp5 1 1.6112 0.1215 13.26 <.0001

Lag Covariance Correlation -1 9 8 7 6 5 4 3 2 1 0 1 2 3 4 5 6 7 8 9 1 

0 0.00553 1 |                    |********************|

1 0.00351 0.634103 |                    |*************       |

Preliminary MSE 0.00331

StandardError

1 -0.634103 0.157839 -4.02

SSE 0.07185893 DFE 24MSE 0.00299 Root MSE 0.05472SBC -73.055453 AIC -76.942964MAE 0.04203192 AICC -75.899486

MAPE 0.28577549 HQC -75.787003Transformed Regression R-Square

0.7313

Total R-Square

0.9401

Order DW1 1.56592 1.456

Standard ApproxError Pr > |t|

Intercept 1 7.8561 0.8502 9.24 <.0001lgdp5 1 1.5203 0.1881 8.08 <.0001

Estimates of Autoregressive ParametersLag Coefficient t Value

Yule-Walker Estimates

Ordinary Least Squares Estimates

Durbin-Watson Statistics

Parameter Estimatest Value

Estimates of Autocorrelations

Variable DF Estimate

Durbin-Watson Statistics

Parameter EstimatesVariable DF Estimate t Value

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 52 of 63Methodology and Data Sources Office of Aviation Policy and Plans

MexicoThe AUTOREG Procedure

Dependent Variable lnPaxMexico

SSE 0.20144725 DFE 25MSE 0.00806 Root MSE 0.08977SBC -49.03339 AIC -51.625063MAE 0.06632491 AICC -51.125063

MAPE 0.39991548 HQC -50.854423Total R-Square 0.9323

Order DW1 0.22792 0.6219

Standard ApproxError Pr > |t|

Intercept 1 9.1779 0.3926 23.38 <.0001lgdp6 1 1.6348 0.0881 18.56 <.0001

Lag Covariance Correlation -1 9 8 7 6 5 4 3 2 1 0 1 2 3 4 5 

6 7 8 9 1 0 0.00746 1 |                    |**

******************|

1 0.00559 0.749384 |                    |***************     |

Preliminary MSE 0.00327

StandardError

1 -0.749384 0.135158 -5.54

SSE 0.05387011 DFE 24MSE 0.00224 Root MSE 0.04738SBC -80.524683 AIC -84.412194MAE 0.03466353 AICC -83.368715

MAPE 0.20972717 HQC -83.256233Transformed

Regression R-Square

0.8764

Total R-Square 0.9819

Order DW1 0.76752 1.3688

Standard ApproxError Pr > |t|

Intercept 1 8.6218 0.6031 14.3 <.0001lgdp6 1 1.7658 0.1354 13.04 <.0001

Yule-Walker Estimates

Durbin-Watson Statistics

Parameter EstimatesVariable DF Estimate t Value

Ordinary Least Squares Estimates

Durbin-Watson Statistics

Parameter EstimatesVariable DF Estimate t Value

Estimates of Autocorrelations

Estimates of Autoregressive ParametersLag Coefficient t Value

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 53 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Other Latin American CountriesThe AUTOREG Procedure

Dependent Variable lnPaxLtnAmOther

SSE 0.04384308 DFE 19MSE 0.00231 Root MSE 0.04804SBC -65.093556 AIC -68.366684MAE 0.03727921 AICC -67.03335

MAPE 0.22136205 HQC -67.595633Total R-Square 0.9694

Order DW1 0.72282 1.2714

Standard ApproxError Pr > |t|

Intercept 1 9.7658 0.3829 25.51 <.0001lgdp3 1 1.5925 0.0898 17.73 <.0001

Post911 1 -0.1231 0.0356 -3.46 0.0026 Post911

Lag Covariance Correlation -1 9 8 7 6 5 4 3 2 1 0 1 2 3 4 5 

6 7 8 9 1 0 0.00199 1 |                    |**

******************|

1 0.00088 0.441731 |                    |*********           |

Preliminary MSE 0.0016

StandardError

1 -0.441731 0.21146 -2.09

SSE 0.03050158 DFE 18MSE 0.00169 Root MSE 0.04116SBC -69.767887 AIC -74.132057MAE 0.02947287 AICC -71.779115

MAPE 0.17523923 HQC -73.103989Transformed

Regression R-Square0.9465

Total R-Square 0.9787

Order DW1 1.02672 1.2968

Standard ApproxError Pr > |t|

Intercept 1 10.0178 0.4607 21.75 <.0001lgdp3 1 1.5325 0.1071 14.32 <.0001

Post911 1 -0.0901 0.0398 -2.27 0.036 Post911

Estimate t Value Variable Label

Estimates of Autoregressive ParametersLag Coefficient t Value

Yule-Walker Estimates

Durbin-Watson Statistics

Parameter EstimatesVariable DF

Estimates of Autocorrelations

Ordinary Least Squares Estimates

Durbin-Watson Statistics

Parameter EstimatesVariable DF Estimate t Value Variable Label

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 54 of 63Methodology and Data Sources Office of Aviation Policy and Plans

ChinaThe REG ProcedureModel: MODEL1Dependent Variable: PaxChina PaxChina

Number of Observations Read 51Number of Observations Used 27Number of Observations with

Missing Values24

Sum of MeanSquares Square

Model 3 1.01E+14 3.37E+13 139.56 <.0001Error 23 5.55E+12 2.41E+11

Corrected Total 26 1.07E+14

Root MSE 491365 R-Square 0.9479Dependent Mean 1806216 Adj R-Sq 0.9411

Coeff Var 27.2041

Parameter StandardEstimate Error

Intercept Intercept 1 -2206081 764579 -2.89 0.0084gdp5 1 80249 5792.32911 13.85 <.0001exch exch 1 -187020 85995 -2.17 0.0402

Post911 Post911 1 -1276985 336376 -3.8 0.0009

Analysis of VarianceSource DF F Value Pr > F

Parameter EstimatesVariable Label DF t Value Pr > |t|

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 55 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Hong KongThe AUTOREG Procedure

Dependent Variable PaxHongKongPaxHongKong

SSE 4.96E+11 DFE 23MSE 2.16E+10 Root MSE 146820SBC 727.912684 AIC 722.729337MAE 102602.061 AICC 724.547518

MAPE 6.15977722 HQC 724.270618Durbin-Watson 0.9226 Total R-Square 0.9658

Standard ApproxError Pr > |t|

Intercept 1 -1504900 183244 -8.21 <.0001gdp3 1 42367 2857 14.83 <.0001SARS 1 -322842 164857 -1.96 0.0624 SARS

Post911 1 -379690 119083 -3.19 0.0041 Post911

Lag Covariance Correlation -1 9 8 7 6 5 4 3 2 1 0 1 2 3 4 5 6 7 8 9 1

 0 1.84E+10 1 |                    |********************|1 7.62E+09 0.414745 |                    |********            |

Preliminary MSE 1.52E+10

StandardError

1 -0.414745 0.193999 -2.14

SSE 3.79E+11 DFE 22MSE 1.72E+10 Root MSE 131278SBC 724.15503 AIC 717.675846MAE 87834.4124 AICC 720.532989

MAPE 5.03771697 HQC 719.602447Durbin-Watson 1.429 Transformed

Regression R-Square

0.9406

Total R-Square 0.9739

Standard ApproxError Pr > |t|

Intercept 1 -1445425 216061 -6.69 <.0001gdp3 1 41236 3211 12.84 <.0001SARS 1 -371918 124323 -2.99 0.0067 SARS

Post911 1 -298789 128451 -2.33 0.0296 Post911

Parameter EstimatesVariable DF Estimate t Value Variable Label

Estimates of Autoregressive ParametersLag Coefficient t Value

Yule-Walker Estimates

Estimates of Autocorrelations

Ordinary Least Squares Estimates

Parameter EstimatesVariable DF Estimate t Value Variable Label

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 56 of 63Methodology and Data Sources Office of Aviation Policy and Plans

IndiaThe AUTOREG Procedure

Dependent Variable PaxIndiaPaxIndia

SSE 1.51E+11 DFE 24MSE 6310016664 Root MSE 79436SBC 692.595962 AIC 688.708451MAE 49607.376 AICC 689.751929

MAPE 12.1308898 HQC 689.864412Total R-Square 0.9533

Order DWThe AUTOREG Procedure

Dependent Variable PaxHongKongPaxHongKong

SSE 4.96E+11 DFE 23MSE 2.16E+10 Root MSE 146820SBC 727.912684 AIC 722.729337MAE 102602.061 AICC 724.547518

MAPE 6.15977722 HQC 724.270618Durbin-Watson 0.9226 Total R-Square 0.9658

Standard ApproxError Pr > |t|

Intercept 1 -1504900 183244 -8.21 <.0001gdp3 1 42367 2857 14.83 <.0001SARS 1 -322842 164857 -1.96 0.0624 SARS

Post911 1 -379690 119083 -3.19 0.0041 Post911

Lag Covariance Correlation -0 1.84E+10 1 |                    |**************

******|1 7.62E+09 0.414745 |                    |********          

  |

Preliminary MSE 1.52E+10

StandardError

1 -0.414745 0.193999 -2.14

SSE 3.79E+11 DFE 22MSE 1.72E+10 Root MSE 131278SBC 724.15503 AIC 717.675846MAE 87834.4124 AICC 720.532989

MAPE 5.03771697 HQC 719.602447Durbin-Watson 1.429 Transformed

Regression R-Square

0.9406

Total R-Square 0.9739

Standard ApproxError Pr > |t|

Intercept 1 -1445425 216061 -6.69 <.0001gdp3 1 41236 3211 12.84 <.0001SARS 1 -371918 124323 -2.99 0.0067 SARS

Post911 1 -298789 128451 -2.33 0.0296 Post911

Parameter EstimatesVariable DF Estimate t Value Variable

Label

Estimates of Autocorrelations

Estimates of Autoregressive ParametersLag Coefficient t Value

Parameter EstimatesVariable DF Estimate t Value Variable

Label

Yule-Walker Estimates

Ordinary Least Squares Estimates

Durbin-Watson Statistics

Ordinary Least Squares Estimates

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 57 of 63Methodology and Data Sources Office of Aviation Policy and Plans

JapanThe AUTOREG Procedure

Dependent Variable lnPaxJapan

SSE 0.076805 DFE 22MSE 0.00349 Root MSE 0.05909SBC -65.1807 AIC -71.659892MAE 0.043651 AICC -68.802749

MAPE 0.268629 HQC -69.733291Total R-Square

0.7968

Order DW1 1.19822 1.7084

Standard ApproxError Pr > |t|

Intercept 1 7.4115 1.1309 6.55 <.0001lgdp2 1 2.1837 0.2547 8.57 <.0001

lnFlatYield 1 -0.282 0.0664 -4.24 0.000311-Sep 1 -0.2069 0.0681 -3.04 0.006 11-SepPost911 1 -0.3793 0.0495 -7.66 <.0001 Post911

Lag Covariance

Correlation -1 9 8 7 6 5 4 3 2 1 0 1 2 3 4 5 

6 7 8 9 1 0 0.00284 1 |                    |**

******************|

1 0.00107 0.377141 |                    |********            |

Preliminary MSE 0.00244

StandardError

1 -0.37714 0.202104 -1.87

SSE 0.063478 DFE 21MSE 0.00302 Root MSE 0.05498SBC -66.8772 AIC -74.652234MAE 0.03966 AICC -70.452234

MAPE 0.244077 HQC -72.340312Transformed

Regression R-Square

0.707

Total R-Square

0.832

Order DW1 1.74262 1.8078

Standard ApproxError Pr > |t|

Intercept 1 7.9347 1.3511 5.87 <.0001lgdp2 1 2.0627 0.3063 6.74 <.0001

lnFlatYield 1 -0.2782 0.088 -3.16 0.004711-Sep 1 -0.2283 0.062 -3.68 0.0014 11-SepPost911 1 -0.3546 0.0612 -5.79 <.0001 Post911

Yule-Walker Estimates

Durbin-Watson Statistics

Parameter EstimatesVariable DF Estimate t Value Variable

Label

Estimates of Autocorrelations

Estimates of Autoregressive ParametersLag Coefficie

ntt Value

Ordinary Least Squares Estimates

Durbin-Watson Statistics

Parameter EstimatesVariable DF Estimate t Value Variable

Label

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 58 of 63Methodology and Data Sources Office of Aviation Policy and Plans

South KoreaThe REG ProcedureModel: MODEL1Dependent Variable: lnPaxSKorea

Number of Observations Read

51

Number of Observations Used

27

Number of Observations with

Missing Values

24

Sum of MeanSquares Square

Model 5 3.6952 0.73904 89.71 <.0001Error 21 0.17301 0.00824

Corrected Total 26 3.86821

Root MSE 0.09077 R-Square 0.9553Dependent Mean 14.90337 Adj R-Sq 0.9446

Coeff Var 0.60903

Parameter StandardEstimate Error

Intercept Intercept 1 6.65152 0.51338 12.96 <.0001lgdp2 1 2.04386 0.13183 15.5 <.0001

11-Sep 11-Sep 1 -0.33107 0.11872 -2.79 0.011Post911 Post911 1 -0.85317 0.09745 -8.75 <.0001

FinanCrisis FinanCrisis 1 -0.24141 0.07804 -3.09 0.0055NWPaxData NWPaxData 1 -0.35373 0.08435 -4.19 0.0004

Label DF t Value Pr > |t|

Analysis of VarianceSource DF F Value Pr > F

Parameter EstimatesVariable

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 59 of 63Methodology and Data Sources Office of Aviation Policy and Plans

TaiwanThe AUTOREG Procedure

Dependent Variable PaxTaiwanPaxTaiwan

SSE 1.46E+12 DFE 23MSE 6.35E+10 Root MSE 252023SBC 757.089559 AIC 751.906211MAE 195807.902 AICC 753.724393

MAPE 12.4557868 HQC 753.447492Total R-Square 0.7615

Order DW1 0.9652 1.5787

Standard ApproxError Pr > |t|

Intercept 1 -1094044 381365 -2.87 0.0087gdp5 1 44253 6086 7.27 <.0001

Post911 1 -740604 199155 -3.72 0.0011 Post911GFC 1 -688658 170952 -4.03 0.0005 GFC

Lag Covariance Correlation -1 9 8 7 6 5 4 3 2 1 0 1 2 3 4 5 6 7 8 9 1 

0 5.41E+10 1 |                    |********************|

1 2.36E+10 0.436802 |                    |*********           |

Preliminary MSE 4.38E+10

StandardError

1 -0.436802 0.191786 -2.28

SSE 1.03E+12 DFE 22MSE 4.70E+10 Root MSE 216829SBC 751.27478 AIC 744.795596MAE 160066.485 AICC 747.652739

MAPE 10.1258116 HQC 746.722197Transformed

Regression R-Square

0.6606

Total R-Square 0.8311

Order DW1 1.23182 1.3034

Standard ApproxError Pr > |t|

Intercept 1 -737410 433592 -1.7 0.1031gdp5 1 37425 6697 5.59 <.0001

Post911 1 -508692 220709 -2.3 0.031 Post911GFC 1 -482119 199003 -2.42 0.0241 GFC

Yule-Walker Estimates

Durbin-Watson Statistics

Parameter EstimatesVariable DF Estimate t Value Variable Label

Estimates of Autocorrelations

Estimates of Autoregressive ParametersLag Coefficient t Value

Ordinary Least Squares Estimates

Durbin-Watson Statistics

Parameter EstimatesVariable DF Estimate t Value Variable Label

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 60 of 63Methodology and Data Sources Office of Aviation Policy and Plans

Rest of Asia PacificThe REG ProcedureModel: MODEL1Dependent Variable: PaxRestofAsiaPac PaxRestofAsiaPac

Number of Observations Read

51

Number of Observations Used

27

Number of Observations with

Missing Values

24

Sum of MeanSquares Square

Model 2 1.24E+13 6.22E+12 128.89 <.0001Error 24 1.16E+12 48225427892

Corrected Total 26 1.36E+13

Root MSE 219603 R-Square 0.9148Dependent Mean 4399320 Adj R-Sq 0.9077

Coeff Var 4.99175

Parameter Standard

Estimate ErrorIntercept Intercept 1 1010236 245150 4.12 0.0004

gdp3 1 45324 3496.75 12.96 <.0001GFC GFC 1 -932356 164114 -5.68 <.0001

Variable Label DF t Value Pr > |t|Parameter Estimates

Analysis of VarianceSource DF F Value Pr > F

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 61 of 63Methodology and Data Sources Office of Aviation Policy and Plans

CanadaThe REG ProcedureModel: MODEL1Dependent Variable: lnPaxCanada lnPaxCanada

Number of Observations Read

62

Number of Observations Used

27

Number of Observations with

Missing Values

35

Sum of MeanSquares Square

Model 3 1.3334 0.44447 230.02 <.0001Error 23 0.04444 0.00193

Corrected Total 26 1.37784

Root MSE 0.04396 R-Square 0.9677Dependent Mean 16.76818 Adj R-Sq 0.9635

Coeff Var 0.26215

Parameter StandardEstimate Error

Intercept Intercept 1 9.36625 0.41105 22.79 <.0001lgdp7 1 1.69743 0.09679 17.54 <.0001

11-Sep 11-Sep 1 -0.14644 0.04996 -2.93 0.0075Post911 Post911 1 -0.27327 0.03915 -6.98 <.0001

Pr > |t|

Analysis of VarianceSource DF F Value Pr > F

Parameter EstimatesVariable Label DF t Value

FAA U.S. Passenger Airline Forecasts Issued on March 24, 2017 Page 62 of 62Methodology and Data Sources Office of Aviation Policy and Plans