a. If you have not looked at a time-series graph of the sales data for Heath Co's line of skiwear (sec data in Exercise 6), do so now. On this plot write a 1 next to the data point for each first quarter, a 2 next to each second quarter, and so forth for all four quarters. Does there appear to be a seasonal pattern in the sales data? Explain why you think the results are as you have found. (c5p6)
b. It seems logical that skiwear would sell better from October through March than from April through September. To test this hypothesis, begin by adding two dummy variables to the data: dummy variable Q2 = 1 for each second quarter (April, May, June) and Q2 = 0 otherwise; dummy variable Q3 = 1 for each third quarter (July, August, September) and Q3 = 0 otherwise. Once the dummy variables have been entered into your data set, estimate the following trend model:
SALES = b0 + b1 (TIME) + b2 Q2 + b3 Q3
SALES = _______ +/- _______ TIME
+/- _______ Q2 +/- _______ Q3
(Circle + or €“ as appropriate for each variable)
Evaluate these results by answering the following:
€¢ Do the signs make sense?
€¢ Are the coefficients statistically different from zero at a 95 percent confidence level (one-tailed test)?
€¢ What percentage of the variation in SALES is explained by this model?
c. Use this model to make a forecast of SALES (SF2) for the four quarters of 2008 and calculate the RMSE for the forecast period.

d. Prepare a time-series plot of SALES (for 1998Q1 through 2007Q4) and SF2 (for 1998Q1 through 2007Q4) to illustrate how SALES and SF2 compare.

SALES ($000) 334,271 328,982 317,921 350,118 Period SF2 ($000) 2008Q1 2008Q2 2008Q3 200804