Performance Lawn Equipment Chapter 8 Case Study:

The following three Parts of this Case Study describe a scenario which Ms. Burke would like your groups to address. Questions 1 and 2 refer to part I, question 3 refers to part II, and question 4 refers to part III.

For each of the four questions, your groups should address each step of the following procedure and provide appropriate analysis within your case study report:

Steps:

1) Plot your data in an appropriate plot and include the plot along with a short discussion in your report. You may ignore this step for Question 3) due to multi-dimensionality under Part II.

2) Fit an appropriate model that gives a sufficiently large R-square and low standard error value. You may want to initially use the "Add Trendline" tool in excel in order to gauge best models. Provide the model in your report.

***Important Note: You should not implement the same type of model, i.e., simple linear regression model, or other model, for all four questions because the same model may not be a good model for different data sets. ***

3) Conduct ANOVA (Analysis of Variance) of the best model and provide the results of ANOVA in your report with some details. For instance, state and critique the R-square and standard error in your report.

4) Check whether the Four Model Assumptions are either met or violated in your model by providing either a figure with descriptions on why any one of the four assumptions is met or violated, or by providing a test result such as the Durbin Watson Test. For the Durbin Watson Test statistics, provide the test statistic, critical values and the decision of whether independence of the residuals is met.

5) Check Model Diagnostics: Determine if each variable and interaction effect (see problem 3.) is significant. All p-values, except the "intercept" p-value, must be below 0.05. If not, you must follow the procedure found on pg. 307 of your textbook, or slide 60 of Chapter 8 PowerPoint.

6) Check for Multicollinearity: Determine if multiple variables explain the same information. If multiple variables explain the same information, consider eliminating one of the variables and re-run regression (fit the new model without the eliminated variable).

7) Make Predictions: Make several future predictions if this is based on time series data or make predictions based on hypothetical (new) observations of your choosing if you have non-time series data.

Part I.

In reviewing the data in the Performance Lawn Equipment Database, Elizabeth Burke noticed that defects received from suppliers have decreased (worksheet Defects After Delivery). Upon investigation, she learned that in 2014, PLE experienced some quality problems due to an increasing number of defects in materials received from suppliers. The company instituted an initiative in August 2015 to work with suppliers to reduce these defects, to more closely coordinate deliveries, and to improve materials quality through reengineering supplier production policies.

Ms. Burke noted that the program appeared to reverse an increasing trend in defects; she would like you to predict what might have happened had the supplier initiative not been implemented (only include Jan. 2014 to Aug. 2015 data for the regression analysis).