+ All Categories
Home > Documents > 20 Lesson Regression Cont

20 Lesson Regression Cont

Date post: 03-Apr-2018
Category:
Upload: umar-ali
View: 222 times
Download: 0 times
Share this document with a friend

of 12

Transcript
  • 7/29/2019 20 Lesson Regression Cont

    1/12

    1

    Slide293NSTAK146B Statistics I 16th April 2010

    Statistics I

    Lesson 20

    Regression cont.

    16th April 2010

  • 7/29/2019 20 Lesson Regression Cont

    2/12

    2

    Slide293NSTAK146B Statistics I 16th April 2010

    Today

    Interpretation of the Regression Coefficients

    How good is your regression model? (Coefficient ofDetermination)

  • 7/29/2019 20 Lesson Regression Cont

    3/12

    3

    Slide293NSTAK146B Statistics I 16th April 2010

    Back to Armands Pizza Parlour

    Interpretation of Coefficients

    Effect of Student Population Size on

    Quarterly Sales for Armand's Pizza

    Parlours

    y = 60 + 5x

    0

    50

    100

    150

    200

    250

    0 5 10 15 20 25 30

    Student Population (000s)

    Qua

    rterlySales(000s)

    Intercept: 60 what doesthis mean?

    Slope: 5 what does

    this mean?

  • 7/29/2019 20 Lesson Regression Cont

    4/12

    4

    Slide293NSTAK146B Statistics I 16th April 2010

    Interpretation of Regression Coefficients

    What about for the Swimmers example?

    And for the Salespeople?

    0 1y b b x

    0 1

    y b b x

  • 7/29/2019 20 Lesson Regression Cont

    5/12

    5

    Slide293NSTAK146B Statistics I 16th April 2010

    Model vs Estimated Simple Linear RegressionEquation

    The estimated simple linear regression equation

    0 1

    y b b x

    y =b0 +b1x +e

    The regression model.

  • 7/29/2019 20 Lesson Regression Cont

    6/12

    6

    Slide293NSTAK146B Statistics I 16th April 2010

    So how good is your Regression Model?

    Coefficient of Determination

    Relationship Among SST, SSR, SSE

    where:

    SST = total sum of squaresSSR = sum of squares due to regression

    SSE = sum of squares due to error

    SST = SSR + SSE

    2( )iy y2

    ( )iy y 2

    ( )i iy y

  • 7/29/2019 20 Lesson Regression Cont

    7/127Slide293NSTAK146B Statistics I 16th April 2010

    Sums of Squares Diagram

    How much ofthe totalvariationfrom theaverage

    (SST) isexplainedby ourregressionmodel

    (SSR) iewhat is theproportionSSR/SST?

  • 7/29/2019 20 Lesson Regression Cont

    8/128Slide293NSTAK146B Statistics I 16th April 2010

    The coefficient of determination is:

    Coefficient of Determination

    where:

    SSR = sum of squares due to regressionSST = total sum of squares

    r2 = SSR/SST

  • 7/29/2019 20 Lesson Regression Cont

    9/129Slide293NSTAK146B Statistics I 16th April 2010

    Coefficient of Determination for Armand

    r2 = SSR/SST = 14200/15730 = .9027

    The regression relationship is very strong;

    90.27% of the variability in sales can be

    explained by the linear relationship betweenthe size of the student population and sales.

  • 7/29/2019 20 Lesson Regression Cont

    10/1210Slide293NSTAK146B Statistics I 16th April 2010

    Sample Correlation Coefficient

    2

    1 )of(sign rbrxy

    ionDeterminatoftCoefficien)of(sign 1brxy

    where:b1 = the slope of the estimated regression

    equation xbby 10

  • 7/29/2019 20 Lesson Regression Cont

    11/1211Slide293NSTAK146B Statistics I 16th April 2010

    2

    1 )of(sign rbrxy

    Armands Sample Correlation Coefficient

  • 7/29/2019 20 Lesson Regression Cont

    12/1212Slide293NSTAK146B Statistics I 16th April 2010

    Over to you

    Now try the example questions


Recommended