Scribd Logo
Upload

Welcome to Scribd!

  • 15 views

    The Sales Manager of An Automobile Parts Dealer Wants To Develop A Model To Predict The Total Annual Sales by Region. Based On Experience It

    Uploaded by

    Juan Jose Rodriguez
    The document discusses developing a multiple linear regression model to predict annual automobile parts sales by region based on April sales data. Several independent variables that may impact sales are identified, including number of retail stores, registered cars, personal income, average car age, and number of sellers. After checking for multicollinearity issues between variables, the regression model is developed eliminating two variables with high multicollinearity. Confidence intervals are then calculated to predict average annual sales at given values of the independent variables.

    Copyright:

    © All Rights Reserved
    Download as XLSX, PDF, TXT or read online from Scribd

    The Sales Manager of An Automobile Parts Dealer Wants To Develop A Model To Predict The Total Annual Sales by Region. Based On Experience It

    Uploaded by

    Juan Jose Rodriguez
    15 views4 pages
    The document discusses developing a multiple linear regression model to predict annual automobile parts sales by region based on April sales data. Several independent variables that may impact sales are identified, including number of retail stores, registered cars, personal income, average car age, and number of sellers. After checking for multicollinearity issues between variables, the regression model is developed eliminating two variables with high multicollinearity. Confidence intervals are then calculated to predict average annual sales at given values of the independent variables.

    Original Title

    H4P2 MLR VIF CONCEPT

    Copyright

    © © All Rights Reserved

    Available Formats

    XLSX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The document discusses developing a multiple linear regression model to predict annual automobile parts sales by region based on April sales data. Several independent variables that may impact sales are identified, including number of retail stores, registered cars, personal income, average car age, and number of sellers. After checking for multicollinearity issues between variables, the regression model is developed eliminating two variables with high multicollinearity. Confidence intervals are then calculated to predict average annual sales at given values of the independent variables.

    Copyright:

    © All Rights Reserved
    Download as XLSX, PDF, TXT or read online from Scribd
    Download as xlsx, pdf, or txt
    15 views4 pages

    The Sales Manager of An Automobile Parts Dealer Wants To Develop A Model To Predict The Total Annual Sales by Region. Based On Experience It

    Uploaded by

    Juan Jose Rodriguez
    The document discusses developing a multiple linear regression model to predict annual automobile parts sales by region based on April sales data. Several independent variables that may impact sales are identified, including number of retail stores, registered cars, personal income, average car age, and number of sellers. After checking for multicollinearity issues between variables, the regression model is developed eliminating two variables with high multicollinearity. Confidence intervals are then calculated to predict average annual sales at given values of the independent variables.

    Copyright:

    © All Rights Reserved
    Download as XLSX, PDF, TXT or read online from Scribd
    Download as xlsx, pdf, or txt
    of 4

    Campus San Luis Potosí, Engineering School

    Course: IN2023:DAE Teacher: Benigno Hernández R.


    Homework #: ______ Period: 2 Delivery date: 9/29/2019
    Homework title: H3P2
    Student(s) List Number

    Note: Students enrolled in this format certifify that 100% of the content of Grade

    1.       The sales manager of an automobile parts dealer wants to develop a model to predict the total annual sales by region. Based on experience it
    can be determined that annual sales estimated based on April are pretty accurate, so that in future years the April forecast could be used to
    review production schedules and have an adequate inventory in retail stores. Several factors are related to sales, including the number of
    retail stores in the region, the number of registered cars in the region until April 1, and the average total personal income, between others.
    A total of five independent variables as the most important (according to the sales manager) was selected. After, he stablished
    these variables, data were collected.
    Annual Number of
    Number of registered Personal Average
    sales (one Number of
    retail cars income age of cars
    hundred sellers
    stores, (thousands) (years)
    thousand)   (10,000)
    Y:AS X1:NRS X2:NRC X3:PI X4:AAC X5:NS
    37.702 39 9.27 85.4 6.5 9
    24.196 21 5.86 60.7 4 5
    32.055 36 8.81 68.1 5.4 7
    3.611 12 3.81 20.2 4 4
    17.625 26 7.31 33.8 3.5 3
    45.919 39 11.62 95.1 6.1 12
    29.6 28 8.96 69.3 4.1 8
    8.114 14 6.28 16.3 2.9 3
    20.116 19 7.77 34.9 5.5 6
    12.994 17 10.92 15.1 3.1 4

    1 1 1 1 1 1 1 1 1 1
    Xt= 39 21 36 12 26 39 28 14 19 17
    9.27 5.86 8.81 3.81 7.31 11.62 8.96 6.28 7.77 10.92
    85.4 60.7 68.1 20.2 33.8 95.1 69.3 16.3 34.9 15.1
    6.5 4 5.4 4 3.5 6.1 4.1 2.9 5.5 3.1
    9 5 7 4 3 12 8 3 6 4

    10 251 80.61 498.9 45.1 61


    251 7229 2162.78 14975.4 1221.4 1750
    A= 80.61 2162.78 700.2045 4335.78 374.511 531.83
    498.9 14975.4 4335.78 32723.95 2515.01 3745.3
    45.1 1221.4 374.511 2515.01 217.95 302.9
    61 1750 531.83 3745.3 302.9 449

    3.92066165 0.08615471 -0.40763431 -0.03076584 -0.9509886 0.51256744


    0.08615471 0.01538438 -0.02825346 -0.00559651 -0.04326289 0.03766787
    A-1= -0.40763431 -0.02825346 0.09128564 0.01149503 0.10730897 -0.11090297
    -0.03076584 -0.00559651 0.01149503 0.00278929 0.0148069 -0.02087866
    -0.9509886 -0.04326289 0.10730897 0.0148069 0.3677537 -0.20088782
    0.51256744 0.03766787 -0.11090297 -0.02087866 -0.20088782 0.22681937

    a) Determine if there are problems of multicollinearity.


    Término
    Constante FIV
    X1:NRS 14.29 Due to four of the VIF's is greater than 5, all the X's can´t be considered as independent variables
    X2:NRC 4.6 therefore the coefficients on the MLRM that relates Y: Annual Sales with these X's can´t be interpreted.
    X3:PI 21.85
    X4:AAC 5.35 There are multicollinearity problems due to VIF's of X1, X3, X4, and X5 are greater than 5
    X5:NS 17.44
    b) Set the multiple linear regression hyperplane eliminating the two X's with more multicollinearity problems.
    EE del tm
    Término Coef Coef. Valor T T Valor p FIV
    Constante -12.28 3.85 -3.19 2.77644511 0.033
    X1:NRS -0.053 0.241 -0.22 2.77644511 0.836 14.29 NO
    X2:NRC 1.857 0.588 3.16 2.77644511 0.034 4.6 YES
    X3:PI 0.407 0.103 3.96 2.77644511 0.017 21.85 YES
    X4:AAC 1.58 1.18 1.34 2.77644511 0.252 5.35 NO
    X5:NS -0.918 0.927 -0.99 2.77644511 0.378 17.44 NO

    c) Set the multiple linear regression hyperplane with all the X's, Interpret b 2. Is the interpretation valid?

    f) Get and interpret an estimate of the variance of b 1.

    EE del
    Término Coef Coef. Valor T Valor p FIV
    Constante -12.28 3.85 -3.19 0.033
    X1:NRS -0.053 0.241 -0.22 0.836 14.29
    X2:NRC 1.857 0.588 3.16 0.034 4.6
    X3:PI 0.407 0.103 3.96 0.017 21.85
    X4:AAC 1.58 1.18 1.34 0.252 5.35
    X5:NS -0.918 0.927 -0.99 0.378 17.44
    VAR(b1)= 0.058081

    h) Obtain and interpret a confidence interval for the average sales at the point [1, 30, 9, 70, 4.5,7]

    Xo[ 1 30 9 70 4.5 7]

    -12.28 Y^= 32.017 1


    -0.053 Sy*x's= 1.94589 Xto= 30
    1.857 t0.025,4= 2.77644511 9
    0.407 v= 4 70
    1.58 Raiz= 0.5672178 4.5
    -0.918 7

    p( 28.9525169 < µY < 35.0814831 )=0.95

    With a confidence level of 95% it is expected that the average annual sales will be at least
    28.95 U and at most 35.08 U when X1: NRS=30, X2: NRC=9, X3: PI=70, X4: AAC=4.5 and
    X5: NS=7.

    You might also like