Assume for now that your model is a valid one. Test at the 5% level of significance whether beer sales exhibit a quarterlyfluctuation. State the method you use and show your calculations. Data and regression results below.

Model - Sales=4421.37+78.18*Temp+2.58*Sun-879.69*Q2-1551.61*Q3+715.39*Q4

I model. This problem is based on a case involving real data from Tryfos (1998, pp. 467-469). According to Tryfos: To the sales manager of Carlsen's Brewery, a formal model to explain and predict beer sales seemed worth a try..... Carlsen's Brewery is one of the major breweries in Canada, with sales in all parts of the country, but the study itself was to be confined to one metropolitan area. In discussing this assignment, the manager pointed out that weather conditions obviously are responsible for most of the short-run variation in beer consumption. "When it is hot", the manager said, "people drink more - it's that simple." This was also the reason for confining the study to one area; since weather conditions vary so much across the country, there was no point in developing a single, countrywide model for beer sales. It was the manager's opinion that a number of models should be developed --one for each major selling area. The available data consisted of 19 consecutive quarterly observations on the following variables: Sales = Quarterly beer sales (in tons) Temp = Average quarterly temperature (in degrees F) Sun = Quarterly total hours of sunlight Q2 = Dummy variable which is 1 for Quarter 2 and 0 otherwise Q3 = Dummy variable which is 1 for Quarter 3 and 0 otherwise Q4 = Dummy variable which is 1 for Quarter 4 and 0 otherwise. The data can be found on the book web site in the file CarlsenQ.txt. Develop a model which can be used to predict quarterly beer sales. Describe any weaknesses in your model. Write up the results in the form of a report that is to be given to the manager at Carlsen's brewery