Business Forecasting

Only available on StudyMode
  • Topic: Forecasting, Exponential smoothing, Time series
  • Pages : 20 (6828 words )
  • Download(s) : 78
  • Published : April 10, 2013
Open Document
Text Preview
Business Forecasting

Contents
1.0 Executive summary…………………………………………………………………………………4 2.0 Introduction……………………………………………………………………………………………5 3.0 Question 1……………………………………………………………………………………………...6 4.1 a) Time series plot…………………………………………………………………………6 4.2 b) Exponential smoothing methods……………………………………………….8 4.3 c) 8 months Forecasted period……………………………………………………11 4.4 d) Forecasting report……………………………………………………………………13 4.0 Question 2…………………………………………………………………………………………….15 5.5 Forecasting in Marketing……………………………………………………………..15 5.6 Financial Forecasting……………………………………………………………………17 5.0 Conclusion…………………………………………………………………………………………….18 6.0 Recommendations………………………………………………………………………………..18 7.0 Appendix………………………………………………………………………………………………19 8.0 References………………………………………………………………………………………….25

1.0 Executive summary
The Holt-Winters multiplicative model proved to be the most applicable method to use to forecast the data for the warehouse store sales (in million dollars) due to it being able to take into account trends and seasonality, which the other methods were not able to show. Unlike moving averages models, the data was non-stationary so the Holt-winters method proved to be the most appropriate exponential smoothing method. The trial and error approaches further validated the compatibility of the Holt-winters multiplicative method through producing the lowest accuracy measures, particularly the lowest MAPE. The actual results showed that 6 of the 8 months produced lower sales than the forecast had predicted, hence results show that the forecasts were based on more uncertainty rather than more accurate predictions. Nevertheless, companies require forecasts in order to plan ahead and make predictions. Even though introducing qualitative approaches will help produce more accurate forecasts but the method of predicting figures cannot be dismissed that easily. A marketing department can conduct both quantitative and qualitative forecasts, by using sales forecasting by estimating figures and using market research and initiating opinion polls to gain the most accurate forecasts. A finance department can use cash flow forecasting to control their levels of costs to gain the most amounts of profits, and maintaining accurate bookkeeping and regular checking of balance sheets and profit-and-loss statements in order to avoid any further inaccuracies which can contribute to forecast better in the future.

2.0 Introduction
The most accurate forecasting techniques relating to monthly warehouse store sales (in million dollars) in the USA from January 1992 to June 2004 will be explored in this assignment. In order to choose the most suitable exponential smoothing method for a time series data, it is important to observe the type of data pattern as shown in the time series plot, such as whether there is a horizontal, trend, seasonal or cyclical pattern (Hanke, 2009). When deciding on the most suitable exponential smoothing method, the model which produces the lowest accuracy measures will be chosen through the help of trial and error. The models that will be tested would include the Simple Exponential smoothing method, Double Exponential smoothing and the Holts Winter method. The exponential smoothing model will be selected on the basis of using the MAPE (Mean Absolute Percentage Error) as the main measure of accuracy. The MAPE is the mean or average of the absolute percentage errors of forecasts. Error is defined as actual or observed value minus the forecasted value. Percentage errors are summed without regard to sign to compute MAPE. This measure is easy to understand because it provides the error in terms of percentages. Also, because absolute percentage errors are used, the problem of cancelling each other out is avoided. Consequently, MAPE has managerial appeal and is a measure commonly used in forecasting and the smaller the MAPE the better the...
tracking img