The purpose of this project is to construct a model to attempt to forecast the interim earnings of Harley Davidson. Harley Davidson, Inc. manufactures premium motorcycles, recreational vehicles, specialized commercial vehicles and parts and accessories. The Company's Motorcycle Division manufactures and markets heavyweight touring and custom motorcycles and a broad range of related products including riding apparel, motorcycle accessories and replacement parts. My purpose is to construct a model using six years worth of historical data and try to predict the earnings for the following year. I will use the interim earnings for the years 1989-1994, and divide them into 24 equal quarters. The model uses a quantative forecasting method called the Multiplicative Time Series Model. It is based on the assumption that there are four factors that have influenced the time series in the past. They are:
These factors will also be influences in the forecast year as well.
If I isolate them I will be able to forecast the earnings for 1995. Finally, I will measure the accuracy of my forecast in order to decide if the model is capable of giving accurate forecasts. The actual time series are expressed in this model as the product of the four components: Yt = T*C*S*I.
During the years of 1989 - 1994, Harley Davidson, Inc. experienced an overall increase in interim earnings. The peak of their earnings occurred in the second quarter of 1994 with a .63; while the lowest earnings occurred during the fourth quarter of 1991 with only a .09.
Chart 2 shows the Center Moving Average and Harley Davidson's actual interim earnings. The CMAV is the product of two factors, Cycle and Trend. It represents the values of interim earnings if Trend and...