|A firm, which has been in existence for some time will have accumulated/ collected considerable data on sales pertaining to different time periods. Such data when arranged chronologically yield ‘time series’. The time series relating to sales represent the past pattern of effective demand for a particular product. Such data can be presented either in a tabular form or graphically for further analysis. The most|
The trend line is then projected into the future by extrapolation. The basic assumption of the trend method is that the past rate of change of the variable under study will continue in the future. This technique yields acceptable results so long as the time series shows a persistent tendency to move in the same direction. Whenever a turning point occurs, the trend projection breaks down. Nevertheless, a forecaster could normally expect to be right in most forecasts especially if the turning points are few and spaced at long intervals from each other.
The real challenge of forecasting is in the prediction of turning points rather than in the projection of trends. It is when turning points occur that management will have to alter and revise its sales and production strategies most drastically.
|There are primarily four sets of factors, which are responsible for the characterization of time series by fluctuations and turning points in a time series, trend, seasonal variations, cyclical fluctuations, and irregular or random forces. The problem in forecasting is separate and measures each of these four factors.|
The fundamental approach is to treat the original time series data (O or observed data) as composed of four parts: a secular trend (T), a seasonal factor (S), a cyclical element (C) and on irregular movement (I). It is generally assumed that these elements are bound together in a multiplicative relationship presented by the equation O = TSCI.
The usual practice is to first compute the trend from the original data. The trend values are then eliminated from observed data (TSCI/T). The next step is to calculate the seasonal index, which is used to remove the seasonal effect (SCI/S). A cycle is then fitted to the remainder, which also contains the irregular effect.
The foregoing approach t the decomposition of time series data is a useful analytical device for understanding the nature of business fluctuations. The trend and seasonal factor can be forecast, but the prediction of cycles is hazardous for the simple reason that there is no regularity in the cyclical behavior.
Though, there are two assumptions underlying this approach:
- The analysis of movements would be in the order of trend, seasonal variation and cyclical charges, and
- The effects of each component are independent of each other.
- See if a relationship exists between the demand for a product and certain economic indicators.
- Establish the relationship through the method of least squares and derive the regression equation. Assuming the relationship to be linear, the equation will be of the form y= a + bx. There can be curve-linear relationships as well.
- Once regression equation is derived, the value of Y i.e. demands, can be estimated for any given value of X.
- Past relationships may not recur. Hence the need for value judgment as well. New factors may also have to be taken into consideration.
- The trend method is based on least square principle of demand forecasting is quite popular due to simplicity.
- It provides good result, which is particularly suitable for long run.
- It is very much simple in the sense that it doesn’t require the knowledge of economic theory and market structure.
- This method is based on the assumption that future events will follow the same path, which may not be true for every time.
- It is not suitable for short-term demand forecasting. This method cannot usually explain the turning points of the business cycle.