# week 1 assignment predictive analytics techniques 1

The purpose of this assignment is to use predictive analytics techniques and the graphs and charts associated with these techniques to forecast outcomes and make business decisions.

Using specified data files, chapter example files, and templates from the “Topic 1 Student Data, Template, and Example Files” topic material, complete Chapter 12 Problems 28, 30, 34, 46, 60, 62 (b through e), and 63 from the textbook. Use the Palisade DecisionTools Excel software to complete these problems where requested and applicable. Problem 60 should be completed using only Excel.

To receive full credit on the assignment, complete the following.

1. Ensure that the Palisade software output is included with your submission.
2. Ensure that Excel files include the associated cell functions and/or formulas if functions and/or formulas are used.
3. Include a written response to all narrative questions presented in the problem by placing it in the associated Excel file.
4. Place each problem in its own Excel file. Ensure that your first and last name are in your Excel file names.

Chapter 12 Problems 28, 30, 34, 46, 60, 62 (b through e), and 63 from the textbook:

28. The file P12_10.xlsx contains annual revenues for a convenience store. If you want to forecast revenue for the next few years with the moving averages method, what span should you use? Will any span work well?.

30. The file P02_28.xlsx contains total monthly U.S. retail sales data. While holding out the final six months of observations for validation purposes, use the method of moving averages with one or more spans of your choice to forecast U.S. retail sales for the next 12 months. Comment on the performance of your model. What makes this time series more challenging to forecast?

34. Consider the American Express closing price data in the file P12_16.xlsx.
a. Create a time series chart of the data. Based on what you see, which of the exponential smoothing models do you think should be used for forecasting? Why?
b. Use Holt’s exponential smoothing to forecast these data, using no holdout period and requesting 20 days of future forecasts. Use the default smoothing constants of 0.1.
c. Repeat part b, optimizing the smoothing constants. Does it make much of an improvement?
d. Repeat parts a and b, this time using a holdout period of 50 days.
e. Write a short report to summarize your results.

46. The file P12_46.xlsx contains monthly time series data for total U.S. retail sales of building materials, garden equipment, and supplies dealers.
a. Is seasonality present in these data? If so, characterize the seasonality pattern.
b. Use the Deseasonalize option in StatTools to forecast the deseasonalized data for each month of the next year using the moving average method with an appropriate span.
c. Does Holt’s exponential smoothing method, with optimal smoothing constants, outperform the moving average method employed in part b? Demonstrate why or why not.

60. The file P12_60.xlsx lists annual revenues (in millions of dollars) for Nike. Create a time series graph of these data. Then superimpose a trend line with Excel’s Trendline option. Which of the possible Trendline options seems to provide the best fit? Using this option, what are your forecasts for the next two years?

62. The file P12_62.xlsx contains data on a motel chain’s revenue and advertising.
b. Use simple exponential smoothing to make predictions for the motel chain’s revenues during the next four quarters.
c. Use Holt’s method to make forecasts for the motel chain’s revenues during the next four quarters.
d. Use Winters’ method to determine predictions for the motel chain’s revenues during the next four quarters.
e. Which of these forecasting methods would you expect to be the most accurate for these data?

63. The file P12_63.xlsx contains data on monthly U.S. permits for new housing units (in thousands of houses).
a. Using Winters’ method, find values of , , and that yield an RMSE as small as possible. Does this method track the housing crash in recent years?
b. Although we have not discussed autocorrelation for smoothing methods, good forecasts derived from smoothing methods should exhibit no substantial autocorrelation in their forecast errors. Is this true for the forecasts in part a?
c. At the end of the observed period, what is the forecast of housing sales during the next few months?

PLEASE USE THE ATTACHED “MIS-665-RS-Topic-1-Student-Data-Template-and-Example-Files.zip” TO GAIN THE DATA FOR EACH QUESTION.