Scribd Logo
Upload

Welcome to Scribd!

  • 87 views

    A10ECMT1020 Ch6 Homework

    Uploaded by

    Thefacelessblog Sydney
    This document provides instructions for a homework assignment on multiple regression analysis using mobile phone quality and price data. Students are asked to: 1) Perform a multiple regression on overall rating scores using various quality ratings as predictors and analyze the results. 2) Use stepwise regression to identify the best predictor variables and test if the full model can be reduced. 3) Compare the full and reduced models based on R-squared, adjusted R-squared, MSE and error standard deviation. The homework aims to help students understand how to apply multiple regression techniques and select the best predictor variables from real-world data.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd

    A10ECMT1020 Ch6 Homework

    Uploaded by

    Thefacelessblog Sydney
    87 views2 pages
    This document provides instructions for a homework assignment on multiple regression analysis using mobile phone quality and price data. Students are asked to: 1) Perform a multiple regression on overall rating scores using various quality ratings as predictors and analyze the results. 2) Use stepwise regression to identify the best predictor variables and test if the full model can be reduced. 3) Compare the full and reduced models based on R-squared, adjusted R-squared, MSE and error standard deviation. The homework aims to help students understand how to apply multiple regression techniques and select the best predictor variables from real-world data.

    Original Title

    A10ECMT1020-Ch6-homework

    Copyright

    © Attribution Non-Commercial (BY-NC)

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document provides instructions for a homework assignment on multiple regression analysis using mobile phone quality and price data. Students are asked to: 1) Perform a multiple regression on overall rating scores using various quality ratings as predictors and analyze the results. 2) Use stepwise regression to identify the best predictor variables and test if the full model can be reduced. 3) Compare the full and reduced models based on R-squared, adjusted R-squared, MSE and error standard deviation. The homework aims to help students understand how to apply multiple regression techniques and select the best predictor variables from real-world data.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    87 views2 pages

    A10ECMT1020 Ch6 Homework

    Uploaded by

    Thefacelessblog Sydney
    This document provides instructions for a homework assignment on multiple regression analysis using mobile phone quality and price data. Students are asked to: 1) Perform a multiple regression on overall rating scores using various quality ratings as predictors and analyze the results. 2) Use stepwise regression to identify the best predictor variables and test if the full model can be reduced. 3) Compare the full and reduced models based on R-squared, adjusted R-squared, MSE and error standard deviation. The homework aims to help students understand how to apply multiple regression techniques and select the best predictor variables from real-world data.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 2

    Faculty of Economics and Business

    Discipline of Operations Management & Econometrics


    Ch.6 Multiple Regression Analysis II

    ECMT 1020: Homework Week 6

    This is not an online homework. You must attempt this question and show your work (including
    computer output from Excel/KaddStat) to your tutor during a workshop on 19 April, 21 April or 22
    April. Your tutor will give you 1 mark for completing most questions, ½ mark for partial attempt and 0
    mark for little or no attempt of the questions.

    Case Study: Mobile Phone Quality and Price

    Reference: Black p.640 –642. The dataset can be found in a worksheet of the file “Chapter 6.xls”
    which can be downloaded from the Blackboard.

    Today, mobile phone is a fashion accessory or a “must have” item. Customers are demanding for a
    range of attributes such as build-in camera, GPS navigation, compatibility with the 3G network, a
    colour display, build-in MP3 player, SMS service, etc. In early 2008, an extensive survey on mobile
    phones was conducted by CHOICE magazine. Mobile phones of various makes and models were
    given an overall rating score, as well as rating scores for numerous categories. Decision makers and
    marketers at mobile-phone manufacturers would like to know what features of a mobile phone are
    important to customers. This would be especially important in helping to design effective marketing
    and advertising campaigns. The rating scores are
    Y = Overall rating score (%)
    X1 = Battery rating score (%)
    X2 = Music rating score (%)
    X3 = SMS rating score (%)

    X4 = Portability rating score (%)


    X5 = Picture quality rating score (%)

    X6 = 3G compatibility (Yes = “1”; No = “0”)


    The make and model of 21 mobile phones surveyed by CHOICE magazine and their rating scores are
    displayed on p.641 in Black et al. (2010).

    ECMT 1020: Business & Economics Statistics B (Autumn 2010) p.1 Dr Boris Choy
    Faculty of Economics and Business
    Discipline of Operations Management & Econometrics
    Ch.6 Multiple Regression Analysis II

    1. Perform a multiple regression analysis by regressing Y on X 1 ,…, X 6 .


    (i) Display the computer output from Excel/KaddStat.
    (ii) Write down the estimated regression models for the mobile phone with and without the
    3G compatibility, respectively.
    (iii) If all rating scores are kept constant, how much can a 3G compatibility mobile phone
    add to the overall rating score?
    (iv) State the null and alternative hypotheses of the overall significance test. Construct the
    ANOVA table and draw your conclusion based on the p-value obtained from KaddStat.
    (v) Predict the overall rating score and its 95% CI of a 3G non-compatible mobile phone
    with a battery score of 58%, music score of 70%, SMS score of 50, portability score of
    65 and picture quality score of 70%. Hint: Use KaddStat to do this.

    2. Use stepwise regression method to find the best subset of predictor variables.
    (i) Display the computer output from Excel/KaddStat.
    (ii) Identify the predictor variables that are excluded from the regression model.
    (iii) Write down the estimated regression equation of the best model.
    (iv) What is the correlation between the observed and predicted y-values?
    (v) Perform a formal statistical test to confirm that the full model in (1) can be reduced to
    the reduced model in (2). Use   0.05. You must give the observed F-value, the
    critical F-value, the decision rule and your conclusion.

    3. Use the computer outputs from the full and reduced models to answer the following questions.
    (i) Is the reduced model better than the full model according to the R 2 ? If not, why?
    (ii) Is the reduced model better than the full model according to the adjusted R 2 ?
    (iii) Is the reduced model better than the full model according to the MSE or the estimate of
    the error standard deviation?
    (iv) How much of the total variation of Y can be explained by the unimportant predictor
    variables in 2(ii)? How many percentages is this?

    *** After attempting these questions, you should have a very good understanding of the topics ***

    ECMT 1020: Business & Economics Statistics B (Autumn 2010) p.2 Dr Boris Choy

    You might also like