# Question: Winters method assumes a multiplicative seasonality but an additive trend

Winters’ method assumes a multiplicative seasonality but an additive trend. For example, a trend of 5 means that the level will increase by five units per period. Suppose that there is actually a multiplicative trend. Then (ignoring seasonality) if the current estimate of the level is 50 and the current estimate of the trend is 1.2, the forecast of demand increases by 20% per period. So the forecast demand for the next period is 50(1.2) and forecast demand for two periods in the future is 50(1.2)2. If you want to use a multiplicative trend in Winters’ method, you should use the following equations (assuming a period is a month):

a. What should (I) and (II) be?

b. Suppose that you are working with monthly data and month 12 is December, month 13 is January, and so on. Also, suppose that L12 = 100, T12 = 1.2, S1 = 0.90, S2 = 0.70, and S3 = 0.95. If you have just observed Y13 = 200, what is the forecast for Y15 using α = β = γ = 0.5 and a multiplicative trend?

a. What should (I) and (II) be?

b. Suppose that you are working with monthly data and month 12 is December, month 13 is January, and so on. Also, suppose that L12 = 100, T12 = 1.2, S1 = 0.90, S2 = 0.70, and S3 = 0.95. If you have just observed Y13 = 200, what is the forecast for Y15 using α = β = γ = 0.5 and a multiplicative trend?

**View Solution:**## Answer to relevant Questions

Consider the file S12_59.xlsx, which contains total monthly U.S. retail sales data. Does a regression approach for estimating seasonality provide forecasts that are as accurate as those provided by (a)Winters’ method and ...The file S11_44.xlsx contains data on pork sales. Price is in dollars per hundred pounds sold, quantity sold is in billions of pounds, per capita income is in dollars, U.S. population is in millions, and GDP is in billions ...The file S12_79.xlsx contains data on mass layoff events in all industries in the U.S. There are two versions of the data: non-seasonally adjusted and seasonally adjusted. Presumably, seasonal factors can be found by ...There are five corner points in the feasible region for the product mix problem. We identified the coordinates of one of them: (560, 1200). Identify the coordinates of the others.a. Only one of these other corner points has ...Suppose that you want to increase all three of the resource availabilities in the product mix model simultaneously by the same percentage. You want this percentage to vary from -25% to 50% in increments of 5%. Modify the ...Post your question