seasonal analysis

16

Upload: hayes

Post on 03-Feb-2016

36 views

Category:

Documents


0 download

DESCRIPTION

Seasonal Analysis. What is the period?. R-code. par(bg='yellow') plot(co2,lwd=2,col=4,xlab='Time, years',ylab='CO2 in parts per million (ppm)', main='Mauna Loa Atmospheric CO2 Concentration') abline(v=seq(1950,2006,by=10),lty=2) abline(h=seq(320,360,by=10),lty=2). R-code. - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: Seasonal Analysis
Page 2: Seasonal Analysis

Mauna Loa Atmospheric CO2 Concentration

Time, years

CO

2 in

par

ts p

er m

illio

n (p

pm)

1960 1970 1980 1990

320

330

340

350

360

What is the period?

par(bg='yellow')plot(co2,lwd=2,col=4,xlab='Time, years',ylab='CO2 in parts per million (ppm)',main='Mauna Loa Atmospheric CO2 Concentration')abline(v=seq(1950,2006,by=10),lty=2)abline(h=seq(320,360,by=10),lty=2)

R-code

Page 3: Seasonal Analysis

Mauna Loa Atmospheric CO2 Concentration

Time, years

CO

2 in

par

ts p

er m

illio

n (p

pm)

1960 1970 1980 1990

320

330

340

350

360

# This line adds yearly gridabline(v=seq(1950,2006,by=1),lty=2)

R-code

Page 4: Seasonal Analysis

# Linear trendtime<-time(co2)l<-lsfit(time,co2)points(time,time*l$coefficients[2]+l$coefficients[1],type='l',col='green')

R-code

Mauna Loa Atmospheric CO2 Concentration

Time, years

CO

2 in

pa

rts

pe

r m

illio

n (

pp

m)

1960 1970 1980 1990

32

03

40

36

0

Page 5: Seasonal Analysis

Mauna Loa Atmospheric CO2 Concentration

Time, years

CO

2 in

par

ts p

er m

illio

n (p

pm)

1960 1970 1980 1990

320

330

340

350

360

# Filter with 1-year window and plot a trendf<-filter(co2,rep(1,12)/12)points(f,col=2,type='l',lwd=2)

R-code

Page 6: Seasonal Analysis

# Plot series minus trendplot(co2-f,ldw=2,col=4,lwd=2,xlab='Time, years',ylab='CO2-trend',main='Detrended CO2 series')abline(h=seq(-4,4,by=2),lty=2)abline(v=seq(1950,2006,by=1),lty=2)

R-code

Detrended CO2 series

Time, years

CO

2-tr

end

1960 1970 1980 1990

-4-2

02

4

Page 7: Seasonal Analysis

UNR * STAT 758 * Fall 2006

# Perform seasonal analysis and plot season componentm<-decompose(co2)plot(m$figure,lwd=2,type='p',pch=19,col=4,xlab='Month',ylab='Seasonal component')axis(1,at=seq(1,12))grid(col='black',lty=2)

R-code

2 4 6 8 10 12

-3-2

-10

12

3

Seasonal Component of CO2 data

Month

Sea

sona

l com

pone

nt

1 2 3 4 5 6 7 8 9 10 11 12

Page 8: Seasonal Analysis

320

330

340

350

360

obse

rved

320

330

340

350

360

tren

d

-3-2

-10

12

seas

onal

-0.5

0.0

0.5

1960 1970 1980 1990

rand

om

Time

Decomposition of additive time series

m<-decompose(co2)plot(m)

R-code

Page 9: Seasonal Analysis
Page 10: Seasonal Analysis

0 2 4 6 8 10 12

-10

-50

510

Data from a Signal+Noise model

Time

Obs

erva

tions

Page 11: Seasonal Analysis

0 2 4 6 8 10 12

-10

-50

510

Data from a Signal+Noise model

Time

Obs

erva

tions

0 2 4 6 8 10 12-3

-2-1

01

23

Data from a Signal+Noise model

Time

Obs

erva

tions

Estimated signal

Page 12: Seasonal Analysis
Page 13: Seasonal Analysis

Mauna Loa Atmospheric CO2 Concentration

Time, years

CO

2 in

par

ts p

er m

illio

n (p

pm)

1960 1970 1980 1990

320

330

340

350

360

Page 14: Seasonal Analysis

12th-Order Difference of CO2 series

Time, years

diff

(CO

2,12

)

1960 1970 1980 1990

0.0

0.5

1.0

1.5

2.0

2.5

3.0

Page 15: Seasonal Analysis

First-Order Difference of CO2 series

Time, years

diff

(CO

2,1)

1960 1970 1980 1990

-2-1

01

2

Page 16: Seasonal Analysis

(12+1)-Difference of CO2 series

Time, years

diff

(diff

(CO

2,12

),1)

1960 1970 1980 1990

-1.0

-0.5

0.0

0.5

1.0