Ch18 Multiple Regression

Embed Size (px)

Citation preview

  • 8/12/2019 Ch18 Multiple Regression

    1/51

    1

    MultipleRegression

    Chapter 17

  • 8/12/2019 Ch18 Multiple Regression

    2/51

    2

    Introduction

    In this chapter we extend the simple linearregression model, and allow for any number of

    independent variables. We expect to build a model that fits the data

    better than the simple linear regression model.

  • 8/12/2019 Ch18 Multiple Regression

    3/51

    3

    Weight

    Calories consumed

    Introduction We all believe that weight is affected by the amount of calories

    consumed. Yet, the actual effect is different from one individual toanother.

    Therefore, a simple linear relationship leaves much unexplained error.

  • 8/12/2019 Ch18 Multiple Regression

    4/51

    4

    Weight

    Calories consumed

    Introduction

    Click to to continue

    In an attempt to reduce the unexplained errors, well adda second explanatory (independent) variable

  • 8/12/2019 Ch18 Multiple Regression

    5/51

    5

    Weight

    Calories consumed

    Weight = b0+ b1Calories+ b2Height + e

    Introduction

    If we believe a persons height explains his/her weight too, we can add this

    variable to our model. The resulting Multiple regression model is shown:

  • 8/12/2019 Ch18 Multiple Regression

    6/51

    6

    We shall use computer printout to

    Assess the model

    How well it fits the data Is it useful

    Are any required conditions violated?

    Employ the model

    Interpreting the coefficients Making predictions using the prediction equation

    Estimating the expected value of the dependent variable

    Introduction

  • 8/12/2019 Ch18 Multiple Regression

    7/51

    7

    Dependent variable Independent variables

    Random error variable

    17.1 Model and Required Conditions

    Coefficients

    We allow k independent variables to potentiallyexplain the dependent variable

    y = b0+ b1x1+ b2x2+ + bkxk+ e

  • 8/12/2019 Ch18 Multiple Regression

    8/51

    8

    The erroreis normally distributed.

    The mean is equal to zero and the standard deviation isconstant (se)for all values of y.

    The errors are independent.

    Model AssumptionsRequired conditions for e

  • 8/12/2019 Ch18 Multiple Regression

    9/51

    9

    If the model assessment indicates good fit to the data, use itto interpret the coefficients and generate predictions.

    Assess the model fit using statistics obtained from the

    sample.

    Diagnose violations of required conditions. Try to remedyproblems when identified.

    17.2 Estimating the Coefficients and

    Assessing the Model The procedure used to perform regression analysis:

    Obtain the model coefficients and statistics using a

    statistical software.

  • 8/12/2019 Ch18 Multiple Regression

    10/51

    10

    Example 1 Where to locate a new motor inn?

    La Quinta Motor Inns is planning an expansion.

    Management wishes to predict which sites are likely to beprofitable.

    Several areas where predictors of profitability can be identifiedare:

    Competition Market awareness

    Demand generators

    Demographics

    Physical quality

    Estimating the Coefficients and

    Assessing the Model, Example

  • 8/12/2019 Ch18 Multiple Regression

    11/51

    11

    Profitability

    Competition Marketawareness Customers Community Physical

    Operating Margin

    Rooms Nearest Officespace

    Enrollment Income Distance

    Distance todowntown.

    Medianhouseholdincome.

    Distance tothe nearestLa Quinta inn.

    Number ofhotels/motelsrooms within3 miles from

    the site.

    X1 x2 x3 x4 x5 x6

    CollegeEnrollment

    Estimating the Coefficients and

    Assessing the Model, Example

  • 8/12/2019 Ch18 Multiple Regression

    12/51

    12

    Data were collected from randomly selected 100 inns that belongto La Quinta, and ran for the following suggested model:

    Margin = b0 b1Rooms b2Nearest b3Officeb4College + b5Income + b6Disttwn + e

    INN MARGIN ROOMS NEAREST OFFICE COLLEGE INCOME DISTTWN1 55.5 3203 4.2 549 8 37 2.7

    2 33.8 2810 2.8 496 17.5 35 14.4

    3 49 2890 2.4 254 20 35 2.6

    4 31.9 3422 3.3 434 15.5 38 12.1

    5 57.4 2687 0.9 678 15.5 42 6.9

    6 49 3759 2.9 635 19 33 10.8

    Estimating the Coefficients and

    Assessing the Model, Example

  • 8/12/2019 Ch18 Multiple Regression

    13/51

    13

    SUMMARY OUTPUT

    Regression Statistics

    Multiple R 0.724611

    R Square 0.525062

    Adjusted 0.49442

    Standard 5.512084

    Observatio 100

    ANOVA

    df SS MS F gnificance F

    Regressio 6 3123.832 520.6387 17.13581 3.03E-13

    Residual 93 2825.626 30.38307

    Total 99 5949.458

    Coeff icient andard Err t Stat P-value Lower 95%Upper 95%

    Intercept 38.13858 6.992948 5.453862 4.04E-07 24.25197 52.02518

    Number -0.00762 0.001255 -6.06871 2.77E-08 -0.01011 -0.00513

    Nearest 1.646237 0.632837 2.601361 0.010803 0.389548 2.902926

    Office Spa 0.019766 0.00341 5.795594 9.24E-08 0.012993 0.026538

    Enrollment 0.211783 0.133428 1.587246 0.115851 -0.05318 0.476744

    Income 0.413122 0.139552 2.960337 0.003899 0.135999 0.690246

    Distance -0.22526 0.178709 -1.26048 0.210651 -0.58014 0.129622

    This is the sample regression equation(sometimes called the prediction equation)

    MARGIN = 38.14 -0.0076ROOMS+1.65NEAREST

    + 0.02OFFICE+0.21COLLEGE+0.41INCOME - 0.23DISTTWN

    Regression Analysis, Excel OutputLa Quinta

    http://localhost/var/www/apps/conversion/tmp/scratch_9/Exercises/La%20Quinta.xlshttp://localhost/var/www/apps/conversion/tmp/scratch_9/Exercises/La%20Quinta.xls
  • 8/12/2019 Ch18 Multiple Regression

    14/51

    14

    Model Assessment -

    Standard Error of Estimate A small value of seindicates (by definition) a small

    variation of the errors around their mean.

    Since the mean is zero, small variation of the errorsmeans the errors are close to zero.

    So we would prefer a model with a small standarddeviation of the error rather than a large one.

    How can we determine whether the standard deviationof the error is small/large?

  • 8/12/2019 Ch18 Multiple Regression

    15/51

    15

    The standard deviation of the error seis estimated bythe Standard Error of Estimate se:

    1kn

    SSEs

    e

    Model Assessment -

    Standard Error of Estimate

    .yThe magnitude of seis judged by comparing it to

  • 8/12/2019 Ch18 Multiple Regression

    16/51

    16

    SUMMARY OUTPUT

    Regression Statistics

    Multiple R 0.724611

    R Square 0.525062

    Adjusted R S 0.49442

    Standard Erro 5.512084

    Observat ions 100

    ANOVA

    df SS MS F gnificance F

    Regression 6 3123.832 520.6387 17.13581 3.03E-13

    Residual 93 2825.626 30.38307

    Total 99 5949.458

    Coefficientsandard Err t Stat P-value Lower 95%Upper 95%

    Intercept 38.13858 6.992948 5.453862 4.04E-07 24.25197 52.02518

    Number -0.00762 0.001255 -6.06871 2.77E-08 -0.01011 -0.00513

    Nearest 1.646237 0.632837 2.601361 0.010803 0.389548 2.902926

    Office Space 0.019766 0.00341 5.795594 9.24E-08 0.012993 0.026538

    Enrollment 0.211783 0.133428 1.587246 0.115851 -0.05318 0.476744

    Income 0.413122 0.139552 2.960337 0.003899 0.135999 0.690246

    Distance -0.22526 0.178709 -1.26048 0.210651 -0.58014 0.129622

    From the printout, se= 5.5121

    Calculating the mean value ofy we have 739.45y

    Standard Error of Estimate

  • 8/12/2019 Ch18 Multiple Regression

    17/51

    17

    Model Assessment

    Coefficient of Determination In our example it seems seis not particularly small, or is it?

    If seis small the model fits the data well, and is considered useful.The usefulness of the model is evaluated by the amount of variability

    in the y values explained by the model. This is done by thecoefficient of determination.

    The coefficient of determination is calculated by

    As you can see, SSE (thus se) effects the value of r2.

    SST

    SSESST

    SST

    SSRR2

  • 8/12/2019 Ch18 Multiple Regression

    18/51

    18

    SUMMARY OUTPUT

    Regression Statistics

    Multiple R 0.724611

    R Square 0.525062

    Adjusted 0.49442

    Standard 5.512084

    Observatio 100

    ANOVA

    df SS MS F gnificance F

    Regressio 6 3123.832 520.6387 17.13581 3.03E-13

    Residual 93 2825.626 30.38307

    Total 99 5949.458

    Coefficient andard Err t Stat P-value Lower 95% pper 95%

    Intercept 72.45461 7.893104 9.179483 1.11E-14 56.78049 88.12874

    ROOMS -0.00762 0.001255 -6.06871 2.77E-08 -0.01011 -0.00513

    NEAREST -1.64624 0.632837 -2.60136 0.010803 -2.90292 -0.38955

    OFFICE 0.019766 0.00341 5.795594 9.24E-08 0.012993 0.026538

    COLLEGE 0.211783 0.133428 1.587246 0.115851 -0.05318 0.476744

    INCOME -0.41312 0.139552 -2.96034 0.003899 -0.69025 -0.136

    DISTTWN 0.225258 0.178709 1.260475 0.210651 -0.12962 0.580138

    Coefficient of Determination

    From the printout, R2= 0.5251that is, 52.51% of the variabilityin the margin values is explainedby this model.

  • 8/12/2019 Ch18 Multiple Regression

    19/51

    19

    To answer the question we test the hypothesis

    H0: b1= b2= = bk= 0H1: At least one biis not equal to zero.

    If at least one biis not equal to zero, the model hassome validity.

    We pose the question:

    Is there at least one independent variable linearlyrelated to the dependent variable?

    Testing the Validity of the Model

  • 8/12/2019 Ch18 Multiple Regression

    20/51

    20

    Note, that if all the data points satisfy the linear equation without errors, yiandcoincide, and thus SSE = 0. In this case all the variation in y is explained bythe regression (SS(Total) = SSR).

    The total variation in y (SS(Total)) can be explained in part by the regression(SSR) while the rest remains unexplained (SSE):SS(Total) = SSR + SSE or

    iy

    2

    ii

    2

    i )y(y)yy( +2

    i )y(y

    If errors exist in small amounts, SSR will be close to SS(Total) and the ratio

    SSR/SSE will be large. This leads to the F ratio test presented next.

    Testing the Validity of the Model

  • 8/12/2019 Ch18 Multiple Regression

    21/51

    21

    Testing for Significance

    1kn

    SSEMSEk

    SSRMSR

    1knSSE

    kSSR

    MSE

    MSRF

    Define the Mean of the Sum of Squares-Regression (MSR)Define the Mean of the Sum of Squares-Error (MSE)

    The ratio MSR/MSE is F-distributed

  • 8/12/2019 Ch18 Multiple Regression

    22/51

    22

    Rejection region

    F>Fa,k,n-k-1

    Testing for Significance

    Note.

    A Large Fresults from a large SSR, which indicates much of the

    variation in y is explained by the regression model; this is when themodel is useful. Hence, the null hypothesis (which states that themodel is not useful) should be rejected when F is sufficiently large.Therefore, the rejection region has the form of F > Fa,k,n-k-1

  • 8/12/2019 Ch18 Multiple Regression

    23/51

    23

    SUMMARY OUTPUT

    Regression Statistics

    Multiple R 0.724611

    R Square 0.525062

    Adjusted 0.49442

    Standard 5.512084Observatio 100

    ANOVA

    df SS MS F gnificance F

    Regressio 6 3123.832 520.6387 17.13581 3.03E-13

    Residual 93 2825.626 30.38307

    Total 99 5949.458

    Coefficient andard Err t Stat P-value ower 95%Upper 95%

    Intercept 72.45461 7.893104 9.179483 1.11E-14 56.78049 88.12874

    ROOMS -0.00762 0.001255 -6.06871 2.77E-08 -0.01011 -0.00513

    NEAREST -1.64624 0.632837 -2.60136 0.010803 -2.90292 -0.38955

    OFFICE 0.019766 0.00341 5.795594 9.24E-08 0.012993 0.026538

    COLLEGE 0.211783 0.133428 1.587246 0.115851 -0.05318 0.476744

    INCOME -0.41312 0.139552 -2.96034 0.003899 -0.69025 -0.136DISTTWN 0.225258 0.178709 1.260475 0.210651 -0.12962 0.580138

    ANOVA

    df SS MS F Significance F

    Regression 6 3123.832 520.6387 17.13581 3.03382E-13

    Residual 93 2825.626 30.38307Total 99 5949.458

    k =

    nk1 =n1 =

    Testing the Model Validity of the La Quinta

    Inns Regression Model

    MSE=SSE/(n-k-1)

    MSR=SSR/k

    MSR/MSE

    SSE

    SSR

    The F ratio test is performed using the ANOVAportion of the regression output

  • 8/12/2019 Ch18 Multiple Regression

    24/51

    24

    ANOVA

    df SS MS F Significance F

    Regression 6 3123.832 520.6387 17.13581 3.03382E-13

    Residual 93 2825.626 30.38307

    Total 99 5949.458

    k =nk1 =

    n1 =

    If alpha = .05, the critical F isFa,k,n-k-1= F0.05,6,100-6-1=2.17

    F = 17.14 > 2.17

    Also, the p-value = 3.033(10)-13.Clearly,p-value=3.033(10)-13

  • 8/12/2019 Ch18 Multiple Regression

    25/51

    25

    b0 = 38.14.This is the y intercept, the value of y when

    all the variables take the value zero. Since the data

    range of all the independent variables do not cover thevalue zero, do not interpret the intercept.

    Interpreting the Coefficients

    Interpreting the coefficients b1through bk

    y = b0+ b1x1+ b2x2++bkxky = b0+ b1(x1+1)+ b2x2++bkxk

    = b0+ b1x1+ b2x2++bkxk + b1

  • 8/12/2019 Ch18 Multiple Regression

    26/51

    26

    Interpreting the Coefficients

    b1=0.0076.In this model, for each additional room

    within 3 mile of the La Quinta inn, the operating margin

    decreases on the average by .0076% (assuming the

    other variables are held constant).

  • 8/12/2019 Ch18 Multiple Regression

    27/51

    27

    b2= 1.65.In this model, for each additional mile that thenearest competitor is to a La Quinta inn, the average operating

    margin increases by 1.65% when the other variables are held

    constant.

    b3 = 0.02. For each additional 1000 sq-ft of office space, theaverage increase in operating margin will be .02%.

    b4= 0.21. For each additional thousand students the averageoperating margin increases by .21% when the othervariables

    remain constant.

    Interpreting the Coefficients

  • 8/12/2019 Ch18 Multiple Regression

    28/51

    28

    b5= 0.41.For additional $1000 increase in medianhousehold income, the average operating marginincreases by .41%, when the other variables remainconstant.

    b6= - 0.23.For each additional mile to the downtown

    center, the average operating margin decreases by

    .23%.

    Interpreting the Coefficients

  • 8/12/2019 Ch18 Multiple Regression

    29/51

    29

    Test statistic

    ib

    ii

    s

    bt

    b d.f. = n - k -1

    Coefficients Standard Error t Stat P-value Lower 95% Upper 95%

    Intercept 38.13858 6.992948 5.453862 4.04E-07 24.25196697 52.02518Number -0.007618 0.00125527 -6.06871 2.77E-08 -0.010110585 -0.00513

    Nearest 1.646237 0.63283691 2.601361 0.010803 0.389548431 2.902926

    Office Spa 0.019766 0.00341044 5.795594 9.24E-08 0.012993078 0.026538

    Enrollment 0.211783 0.13342794 1.587246 0.115851 -0.053178488 0.476744

    Income 0.413122 0.1395524 2.960337 0.003899 0.135998719 0.690246

    Distance -0.225258 0.17870889 -1.26048 0.210651 -0.580138524 0.129622

    The hypothesis for each bi is

    Excel printout

    H0: bi0

    H1: bi0

    Testing the Coefficients

    For example, a test for b1:t = (-.007618-0)/.001255 = -6.068Suppose alpha=.01. t.005,100-6-1=3.39There is sufficient evidence to rejectH

    0at 1% significance level.

    Moreover the p=value of the test is2.77(10-8). Clearly H0is stronglyrejected. The number of rooms islinearly related to the margin.

  • 8/12/2019 Ch18 Multiple Regression

    30/51

    30

    Coefficients Standard Error t Stat P-value Lower 95% Upper 95%

    Intercept 38.13858 6.992948 5.453862 4.04E-07 24.25196697 52.02518Number -0.007618 0.00125527 -6.06871 2.77E-08 -0.010110585 -0.00513

    Nearest 1.646237 0.63283691 2.601361 0.010803 0.389548431 2.902926

    Office Spa 0.019766 0.00341044 5.795594 9.24E-08 0.012993078 0.026538

    Enrollment 0.211783 0.13342794 1.587246 0.115851 -0.053178488 0.476744

    Income 0.413122 0.1395524 2.960337 0.003899 0.135998719 0.690246

    Distance -0.225258 0.17870889 -1.26048 0.210651 -0.580138524 0.129622

    The hypothesis for each bi is

    Excel printout

    H0: bi0

    H1: bi0

    Testing the Coefficients

    See next the interpretationof the p-value results

  • 8/12/2019 Ch18 Multiple Regression

    31/51

    31

    Interpretation

    Interpretation of the regression results for this model

    The number of hotel and motel rooms, distance to thenearest motel, the amount of office space, and the medianhousehold income are linearly related to the operating margin

    Students enrollment and distance from downtown are notlinearly related to the margin

    Preferable locations have only few other motels nearby,much office space, and the surrounding households areaffluent.

  • 8/12/2019 Ch18 Multiple Regression

    32/51

    32

    The model can be used for making predictions by Producing prediction interval estimate of the particular

    value of y, for given values of xi.

    Producing a confidence interval estimate for theexpected value of y, for given values of xi.

    The model can be used to learn aboutrelationships between the independent variables xi,and the dependent variable y, by interpreting thecoefficients bi

    Using the Regression Equation

  • 8/12/2019 Ch18 Multiple Regression

    33/51

    33

    Predict the average operating margin of an inn at a sitewith the following characteristics:

    3815 rooms within 3 miles, Closet competitor 3.4 miles away,

    476,000 sq-ft of office space,

    24,500 college students,

    $39,000 median household income, 3.6 miles distance to downtown center.

    MARGIN = 38.14 -0.0076(3815)-1.646(.9) + 0.02(476)

    +0.212(24.5) -0.413(35) + 0.225(11.2) = 37.1%

    La Quinta

    La Quinta Inns, Predictions

    http://localhost/var/www/apps/conversion/tmp/scratch_9/Exercises/La%20Quinta.xlshttp://f/Study%20Guide/Ch19-Multiple%20Regression/WINDOWS/Desktop/Power%20Point/Keller-Xms/Xm19-01.xlshttp://localhost/var/www/apps/conversion/tmp/scratch_9/Exercises/La%20Quinta.xlshttp://f/Study%20Guide/Ch19-Multiple%20Regression/WINDOWS/Desktop/Power%20Point/Keller-Xms/Xm19-01.xls
  • 8/12/2019 Ch18 Multiple Regression

    34/51

    34

    Interval estimates by Excel (Data analysis plus)

    Prediction Interval

    Margin

    Predicted value = 37.09149

    Prediction Interval

    Lower limit = 25.39527

    Upper limit = 48.78771

    Interval Estimate of Expected Value

    Lower limit = 32.96972

    Upper limit = 41.21326

    It is predicted that the averageoperating margin will lie

    within 25.4% and 48.8%,with 95% confidence.

    It is expected the averageoperating margin of all sitesthat fit this category falls

    within 33% and 41.2% with95% confidence.

    The average inn would not beprofitable (Less than 50%).

    La Quinta Inns, Predictions

  • 8/12/2019 Ch18 Multiple Regression

    35/51

    35

    18.2 Qualitative Independent Variables

    In many real-life situations one or moreindependent variables are qualitative.

    Including qualitative variables in a regressionanalysis model is done via indicator variables.

    An indicator variable (I) can assume one out of

    two values, zero or one.

    1 if a first condition out of two is met0 if a second condition out of two is met

    I=1 if data were collected before 19800 if data were collected after 1980

    1 if the temperature was below 50o0 if the temperature was 50o or more

    1 if a degree earned is in Finance0 if a degree earned is not in Finance

  • 8/12/2019 Ch18 Multiple Regression

    36/51

    36

    Qualitative Independent Variables;

    Example: Auction Car Price (II) Example 2 - continued

    Recall: A car dealer wants to predict the auction

    price of a car. The dealer believes now that both odometer reading

    and car colorare variables that affect a cars price.

    Three color categories are considered: White Silver

    Other colors

    Note: Color is aqualitative variable.

  • 8/12/2019 Ch18 Multiple Regression

    37/51

    37

    Example 2 - continued

    I1 = 1 if the color is white0 if the color is not white

    I2 =1 if the color is silver0 if the color is not silver

    The category Other colors is defined by:

    I1= 0; I2= 0

    Qualitative Independent Variables;

    Example: Auction Car Price (II)

  • 8/12/2019 Ch18 Multiple Regression

    38/51

    38

    Note: To represent the situation of three possiblecolors we need only two indicator variables.

    Generally to represent a nominal variable with m

    possible values, we must create m-1indicatorvariables.

    How Many Indicator Variables?

  • 8/12/2019 Ch18 Multiple Regression

    39/51

    39

    Solution

    the proposed model is

    y = b0+ b1(Odometer) + b2I1+ b3I2+ e The data

    Price Odometer I-1 I-2

    14636 37388 1 0

    14122 44758 1 014016 45833 0 0

    15590 30862 0 0

    15568 31705 0 1

    14718 34010 0 1

    . . . .

    . . . .

    White color

    Other color

    Silver color

    Qualitative Independent Variables;

    Example: Auction Car Price (II)

    Enter the data in Excel as usual

  • 8/12/2019 Ch18 Multiple Regression

    40/51

    40

    Odometer

    Price

    Price = 16.837 - .0591(Odometer) + .0911(0) + .3304(1)

    Price=16.837 - .0591(Odometer) + .0911(1) + .3304(0)

    Price = 16.837 - .0591(Odometer) + .0911(0) + .3304(0)

    From Excel we get the regression equationPRICE = 16.837 - .0591(Odometer) + .0911(I-1) + .3304(I-2)

    Example: Auction Car Price (II)

    The Regression Equation

  • 8/12/2019 Ch18 Multiple Regression

    41/51

    41

    From Excel we get the regression equation

    PRICE = 16701-.0591(Odometer)+.0911(I-1)+.3304(I-2)

    A white car sells, on the average,for $91.1 more than a car of the

    Other color category

    A silver color car sells, on the average,for $330.4 more than a car of theOther color category.

    For one additional mile theauction price decreases by5.91 cents on the average.

    Example: Auction Car Price (II)

    The Regression EquationInterpreting the equation

  • 8/12/2019 Ch18 Multiple Regression

    42/51

    42

    SUMMARY OUTPUT

    Regression Statistics

    Multiple R 0.837135

    R Square 0.700794Adjusted R Square 0.691444

    Standard Error 0.304258

    Observations 100

    ANOVA

    df SS MS F ignificance F

    Regression 3 20.814919 6.938306 74.9498 4.65E-25

    Residual 96 8.8869809 0.092573

    Total 99 29.7019

    Coefficient tandard Err t Stat P-value Lower 95%Upper 95%

    Intercept 16.83725 0.1971054 85.42255 2.28E-92 16.446 17.2285

    Odometer -0.059123 0.0050653 -11.67219 4.04E-20 -0.069177 -0.049068

    I-1 0.091131 0.0728916 1.250224 0.214257 -0.053558 0.235819

    I-2 0.330368 0.0816498 4.046157 0.000105 0.168294 0.492442

    There is insufficient evidenceto infer that a white color car anda car of other color sell for adifferent auction price.

    There is sufficient evidenceto infer that a silver color carsells for a larger price than acar of the other color category.

    Car Price-Dummy

    Example: Auction Car Price (II)

    The Regression Equation

    http://localhost/var/www/apps/conversion/tmp/scratch_9/Exercises/CarPrice%20Dummy.xlshttp://localhost/var/www/apps/conversion/tmp/scratch_9/Exercises/CarPrice%20Dummy.xlshttp://localhost/var/www/apps/conversion/tmp/scratch_9/Exercises/CarPrice%20Dummy.xlshttp://localhost/var/www/apps/conversion/tmp/scratch_9/Exercises/CarPrice%20Dummy.xls
  • 8/12/2019 Ch18 Multiple Regression

    43/51

    43

    Recall: The Dean wanted to evaluate applications for theMBA program by predicting future performance of the

    applicants. The following three predictors were suggested:

    Undergraduate GPA

    GMAT score

    Years of work experience

    It is now believed that the type of undergraduate degreeshould be included in the model.

    Qualitative Independent Variables;

    Example: MBA Program Admission (II)

    Note: The undergraduate

    degree is qualitative.

  • 8/12/2019 Ch18 Multiple Regression

    44/51

    44

    Qualitative Independent Variables;

    Example: MBA Program Admission (II)

    I1 =1 if B.A.0 otherwise

    I2 =1 if B.B.A0 otherwise

    The category Other group is defined by:

    I1= 0; I2= 0; I3= 0

    I3 =1 if B.Sc. or B.Eng.0 otherwise

  • 8/12/2019 Ch18 Multiple Regression

    45/51

    45

    SUMMARY OUTPUT

    Regression Statistics

    Multiple R 0.746053

    R Square 0.556595

    Adjusted R Square 0.524151

    Standard Error 0.729328

    Observations 89

    ANOVA

    df SS MS F gnificance F

    Regression 6 54.75184 9.125307 17.15544 9.59E-13

    Residual 82 43.61738 0.531919

    Total 88 98.36922

    Coeffic ientsandard Err t Stat P-value ower 95%Upper 95%

    Intercept 0.189814 1.406734 0.134932 0.892996 -2.60863 2.988258

    UnderGPA -0.00606 0.113968 -0.05317 0.957728 -0.23278 0.22066

    GMAT 0.012793 0.001356 9.432831 9.92E-15 0.010095 0.015491

    Work 0.098182 0.030323 3.237862 0.001739 0.03786 0.158504

    I-1 -0.34499 0.223728 -1.54199 0.126928 -0.79005 0.100081

    I-2 0.705725 0.240529 2.934058 0.004338 0.227237 1.184213

    I-3 0.034805 0.209401 0.166211 0.8684 -0.38176 0.45137

    Qualitative Independent Variables;

    Example: MBA Program Admission (II)MBA-II

    http://localhost/var/www/apps/conversion/tmp/scratch_9/Exercises/MBA-II.xlshttp://localhost/var/www/apps/conversion/tmp/scratch_9/Exercises/MBA-II.xlshttp://localhost/var/www/apps/conversion/tmp/scratch_9/Exercises/MBA-II.xlshttp://localhost/var/www/apps/conversion/tmp/scratch_9/Exercises/MBA-II.xls
  • 8/12/2019 Ch18 Multiple Regression

    46/51

    46

    Applications in Human Resources

    Management: Pay-Equity Pay-equity can be handled in two different forms:

    Equal pay for equal work

    Equal pay for work of equal value.

    Regression analysis is extensively employed incases of equal pay for equal work.

  • 8/12/2019 Ch18 Multiple Regression

    47/51

    47

    Human Resources Management:

    Pay-Equity Example 3

    Is there sex discrimination against female managers

    in a large firm? A random sample of 100 managers was selectedand data were collected as follows:

    Annual salary

    Years of education Years of experience

    Gender

  • 8/12/2019 Ch18 Multiple Regression

    48/51

    48

    Solution Construct the following multiple regression model:

    y = b0+ b1Education + b2Experience + b3Gender + e Note the nature of the variables:

    Educationquantitative Experiencequantitative

    Genderqualitative (Gender = 1 if male; =0 otherwise).

    Human Resources Management:

    Pay-Equity

  • 8/12/2019 Ch18 Multiple Regression

    49/51

    49

    SolutionContinued (HumanResource)

    Human Resources Management:

    Pay-Equity

    SUMMARY OUTPUT

    Regression Statistics

    Multiple R 0.83256

    R Square 0.693155

    Adjusted R Square 0.683567

    Standard Error 16273.96

    Observations 100

    ANOVA

    df SS MS F gnificance F

    Regression 3 5.74E+10 1.91E+10 72.28735 1.55E-24Residual 96 2.54E+10 2.65E+08

    Total 99 8.29E+10

    Coeff ic ient andard Err t Stat P-value Lower 95%Upper 95%

    Intercept -5835.1 16082.8 -0.36282 0.71754 -37759.2 26089.02

    Education 2118.898 1018.486 2.08044 0.040149 97.21837 4140.578

    Experience 4099.338 317.1936 12.92377 9.89E-23 3469.714 4728.963

    Gender 1850.985 3703.07 0.499851 0.618323 -5499.56 9201.527

    Analysis and Interpretation The model fits the data quite well.

    The model is very useful. Experience is a variable strongly

    related to salary. There is no evidence of sex discrimination.

    http://localhost/var/www/apps/conversion/tmp/scratch_9/Exercises/HumanResource.xlshttp://localhost/var/www/apps/conversion/tmp/scratch_9/Exercises/HumanResource.xls
  • 8/12/2019 Ch18 Multiple Regression

    50/51

    50

    SolutionContinued (HumanResource)

    Human Resources Management:

    Pay-Equity

    SUMMARY OUTPUT

    Regression Statistics

    Multiple R 0.83256

    R Square 0.693155

    Adjusted R Square 0.683567

    Standard Error 16273.96

    Observations 100

    ANOVA

    df SS MS F gnificance F

    Regression 3 5.74E+10 1.91E+10 72.28735 1.55E-24Residual 96 2.54E+10 2.65E+08

    Total 99 8.29E+10

    Coeff ic ient andard Err t Stat P-value Lower 95%Upper 95%

    Intercept -5835.1 16082.8 -0.36282 0.71754 -37759.2 26089.02

    Education 2118.898 1018.486 2.08044 0.040149 97.21837 4140.578

    Experience 4099.338 317.1936 12.92377 9.89E-23 3469.714 4728.963

    Gender 1850.985 3703.07 0.499851 0.618323 -5499.56 9201.527

    Analysis and Interpretation Further studying the data we find:

    Average experience (years) for women is 12.Average experience (years) for men is 17

    Average salary for female manager is $76,189Average salary for male manager is $97,832

    http://localhost/var/www/apps/conversion/tmp/scratch_9/Exercises/HumanResource.xlshttp://localhost/var/www/apps/conversion/tmp/scratch_9/Exercises/HumanResource.xls
  • 8/12/2019 Ch18 Multiple Regression

    51/51

    51

    Review problems

    http://localhost/var/www/apps/conversion/tmp/Additional/Chapter%2019-20.ppthttp://localhost/var/www/apps/conversion/tmp/scratch_9//Dads-moms/OnlineKeller/Review%20Problems/Chapter%2019-20.ppthttp://localhost/var/www/apps/conversion/tmp/scratch_9//Dads-moms/OnlineKeller/Review%20Problems/Chapter%2019-20.ppthttp://localhost/var/www/apps/conversion/tmp/Additional/Chapter%2019-20.ppt