The file P08_30.xlsx contains quarterly revenue for Nike for the years 1991 to 1998. It also contains quarterly “indicator” variables Q1, Q2, and Q3. Here Q1 is 1 for the first quarter of a fiscal...

The file P08_30.xlsx contains quarterly revenue for Nike for the years 1991 to 1998. It also contains quarterly “indicator” variables Q1, Q2, and Q3. Here Q1 is 1 for the first quarter of a fiscal year (July–September) and 0 otherwise. Q2 and Q3 are defined similarly for the second and third quarters of the fiscal year (October–December and January–March). The “Quarter #” variable is simply the chronological number of the quarter, 1 to 32. We would like to build a quantitative model to explain the variation in quarterly revenue. A reasonable model is as follows:




where a, b, c, d, and e are parameters to estimate.

a.  Find the values of a, b, c, d, and e that best fit this model.


b.  What does your model say about the trend and seasonal aspects of Nike sales? (Hint: The trend effect is captured by the term involving Quarter #. Seasonal effects may be interpreted relative to the quarter, Q4, that we have omitted from the analysis.)