k_ppt- simple regression and correlation

Upload: dhara-mehta

Post on 10-Apr-2018

223 views

Category:

Documents


0 download

TRANSCRIPT

  • 8/8/2019 K_PPT- Simple Regression and Correlation

    1/21

    Basic Business Statistics:

    Concepts & Applications

    Simple Linear Regression

    & Correlation

  • 8/8/2019 K_PPT- Simple Regression and Correlation

    2/21

    Regression Analysis

    Managerial decisions often are based on the relationshipbetween two or more variables .For eg Afterconsidering the relationship between advertising

    expenditures and sales ,a marketing manager mightattempt to predict sales for a given level ofadvertising expenditures.

    What is Regression analysis ?

    Regression analysis establishes the nature of therelationship between various variables throughdevelopment of a mathematical function.

  • 8/8/2019 K_PPT- Simple Regression and Correlation

    3/21

    Y Xi i i! F F I0 1

    Linear Regression Model

    Relationship Between Variables Is a

    Linear Function

    Dependent

    (Response)

    Variable

    Independent

    (Explanatory)

    Variable

    SlopeY-InterceptRandom

    Error

  • 8/8/2019 K_PPT- Simple Regression and Correlation

    4/21

    Ii= Random Error

    Y

    X

    PopulationLinear Regression Model

    Observed

    Value

    Observed Value

    Q F FYX iX! 0 1

    Y Xi i i! F F I0 1

  • 8/8/2019 K_PPT- Simple Regression and Correlation

    5/21

    ei= Random

    Error

    Y

    X

    SampleLinear Regression Model

    Observed Value

    Unsampled

    Observation

    Y b b X ei i i! 0 1

    Y b b X i i! 0 1

  • 8/8/2019 K_PPT- Simple Regression and Correlation

    6/21

    Ordinary Least Squares

    1. Best Fit Means Difference Between

    Actual Values (Yi) & Predicted Values

    ( ) Are a Minimum

    But Positive Differences Off-Set Negative

    2. OLS Minimizes the Sum of the

    Squared Differences (or Errors)

    Yi

    Y Yei i

    i

    n

    ii

    n

    !! !

    2

    1

    2

    1

  • 8/8/2019 K_PPT- Simple Regression and Correlation

    7/21

    e2

    Y

    X

    e1 e3

    e4

    Ordinary Least SquaresGraphically

    Y b b X ei i i! 0 1

    Y b b X i i! 0 1

    OLS Minimizes e e e e eii

    n

    2

    1

    1

    222

    3

    24

    2! !

    PredictedValue

  • 8/8/2019 K_PPT- Simple Regression and Correlation

    8/21

    Y b b X

    b

    X Y nXY

    X n X

    b Y b X

    i i

    i i

    i

    n

    ii

    n

    !

    !

    !

    !

    !

    0 1

    11

    2 2

    1

    0 1

    Coefficient Equations

    Sample Regression

    Equation

    Sample Slope

    Sample Y-Intercept

    n = # (Xi, Yi) Pairs

    Average Xis,

    Then Square

  • 8/8/2019 K_PPT- Simple Regression and Correlation

    9/21

    Example (Ad -Sales)

    Youre a marketing analyst for Hasbro

    Toys. You gather the following data:

    Ad Sales (Units)1 1

    2 1

    3 2

    4 25 4

    What is the relationship

    between sales & advertising?

  • 8/8/2019 K_PPT- Simple Regression and Correlation

    10/21

    0

    1

    2

    34

    0 1 2 3 4 5

    Scatter DiagramExample: Ad-Sales

    Sales

    Advertising

  • 8/8/2019 K_PPT- Simple Regression and Correlation

    11/21

    Example (Ad $-Sales)Solution Table

    Xi Yi Xi2

    Yi2

    XiYi

    1 1 1 1 12 1 4 1 2

    3 2 9 4 6

    4 2 16 4 85 4 25 16 20

    15 10 55 26 37

  • 8/8/2019 K_PPT- Simple Regression and Correlation

    12/21

    Example (Ad -Sales)Calculations

    b

    X Y nXY

    X n X

    b Y b X

    Y X

    i i

    i

    n

    i

    i

    n

    i i

    1

    1

    2 2

    1

    0 1

    37 5 3 2

    55 5 0 70

    2 0 70 3 0 10

    0 10 0 70

    !

    !

    !

    ! ! !

    !

    !

    !

    .

    . .

    . .

  • 8/8/2019 K_PPT- Simple Regression and Correlation

    13/21

    Example (Ad -Sales)Interpretation

    1. Slope (b1)

    Sales Volume (Y) Is Expected to Increase

    by 0.7 Units for Each Re1 Increase inAdvertising (X)

    2. Y-Intercept (b0)

    Average Value of Sales Volume (Y

    ) Is-.10 Units When Advertising (X) Is 0

    Difficult to Explain to Marketing Manager

    Expect Some Sales Without Advertising

  • 8/8/2019 K_PPT- Simple Regression and Correlation

    14/21

    Correlation Coefficient

    1. Answer How Strong Is the LinearRelationship Between 2 Variables?

    2. Coefficient of Correlation Used Population correlation coefficient denoted

    V (Rho) Values range from -1 to +1

    Measures degree of association

    3. Used Mainly for Understanding and forinterpretaion Coeff. Of Determination isused.

  • 8/8/2019 K_PPT- Simple Regression and Correlation

    15/21

    Measures of Variationin Regression

    1. Total Sum of Squares (SST)

    Measures variation of observed Yi around

    the mean,DY

    2. Explained Variation (SSR)

    Variation due to relationship between

    X & Y

    3. Unexplained Variation (SSE)

    Variation due to other factors

  • 8/8/2019 K_PPT- Simple Regression and Correlation

    16/21

    Y

    X

    DY

    Xi

    Variation Measures

    Total Sum of

    Squares (Yi- Y)2

    Unexplained Sum of

    Squares (Yi - Yi)2^

    Explained Sum of

    Squares (Yi - Y)2^

    Yi

    Y b b X i i! 0 1

  • 8/8/2019 K_PPT- Simple Regression and Correlation

    17/21

    Proportion of Variation Explained by

    Relationship Between X & Y

    Coefficient of Determination

    0e r2 e 1

    r

    b Y b X Y n Y

    Y n Y

    i i i

    i

    n

    i

    n

    i

    i

    n

    2

    0 1

    2

    11

    2

    1

    2

    ! !

    !

    !!

    !

    Explained Variation

    Total Vari ation

    SSR

    SST

  • 8/8/2019 K_PPT- Simple Regression and Correlation

    18/21

    Interpretation of r square is useful

    Rsquare falls between -1

    and +1

    .It measures the amount of variation which

    is explained by its relation to he

    variable or by the regression line .

  • 8/8/2019 K_PPT- Simple Regression and Correlation

    19/21

    Coefficients of Determination (r2)and Correlation (r)

    r2 = 1, r2 = 1,

    r2 = .8, r2 = 0,Y

    Yi= b

    0+b1Xi

    X

    ^

    Y

    Yi= b

    0+b1Xi

    X

    ^

    Y

    Yi= b

    0+b1Xi

    X

    ^

    Y

    Yi= b

    0+b1Xi

    X

    ^

    r = +1 r = -1

    r = +0.9 r = 0

  • 8/8/2019 K_PPT- Simple Regression and Correlation

    20/21

    Corr Cofficient calculations

    By using the formula r square is found to be equal to0.8167.

    The Coefficient of correlation is - 0.9037. the sign of r is

    determined by the sign of the slope of theregression line .

    81% of the variation in sales is explained by the variationin the advt amount .The remaining 19%of thevariation is due to factors other than Advt Amount

    .the reg line is being evaluated as good and can beused for predictive purposes . Suppose if Rs 6000 isspent on advertising the sales is estimated to be4100 units.

  • 8/8/2019 K_PPT- Simple Regression and Correlation

    21/21

    Pearson Product-Moment Coefficient

    of Correlation

    SampleCoefficient of Correlation

    r

    X X Y Y

    X X Y Y

    i i

    i

    n

    i

    i

    n

    i

    i

    n

    !

    !

    !

    ! !

    1

    2

    1

    2

    1

    Coefficient of Determination