MOVING AVERAGE
As a tool of forecasting, the method of moving averages
attempts to forecast values on the basis of the average of the values of past few periods.Successive values are calculated by considering the new value and dropping the old one.
Example-The demand for an item is observed for 15 months and recorded below
MONTH DEMAND 1 280 2 288 3 266 4 295 5 302 6 310 7 303 8 328 9 309 10 315 11 320 12 332 13 310 14 308 15 320
Calculate (i) 3 monthly (ii) 4 monthly moving averages?
Solution-
Month Demand 3- Monthly moving average 4- Monthly moving average 1 280 2 288 3 266 4 295 278 5 302 283 282.3 6 310 287.7 287.8 7 303 302.3 293.3 8 328 305 302.5 9 309 313.7 310.8 10 315 313.3 312.5 11 320 317.3 313.8 12 332 314.7 318 13 310 322.3 319 14 308 320.7 319.3 15 320 316.7 317.5 16 312.7 317.5
EXPONENTIAL SMOOTHING
This is another time series forecasting technique where the forecast for the next period is calculated as weighted average of all previous values. It is based on the premise that the most recent value is the most important for predicting the future value. Also it presumes that values prior to the current value are also relevant but in declining importance as we go back in time. The weights decline exponentially as we consider the older values.
The choice of smoothing constant,a-
Choosing an appropriate value of the smoothing constant a is an important matter. This is because the choice of this value can make the difference between an accurate and an inaccurate forecast.The difference between an actual value and a forecasted value is called a forecast error.
A measure of overall error of the forecasts made is the Mean Absolute Deviation,MAD.
MAD= ?ÂŚyt-FtÂŚ
n

Example- The demand for a particular item during the…