Exponential Smoothing Method

the most recent demand is given the highest weight a and the weights are decreased by a factor (1 - a) as the data gets older. 

the forecast for period (n + 1) uses the forecast for period n and the actual demand for period n. The value of a is generally chosen between 0.1 and 0.4. In other words, the weights assigned to the actual demand is less than that of the forecasted demand, the reason being, the actual demands fluctuate a lot, while the forecast has smoothed the fluctuations. 

Using the sample data given in Table 2.1, determine the forecast for month 7 using the Exponential Smoothing method. 

In order to use this method, the value of a and the initial forecast for month 1 (Fj) are necessary. The value of can be chosen by averaging all the demands or using the most recent demand value. Assuming a = 0.2, Fj = 307 (Averaging method) and using Equation 2.4, we get the following  

NOTE The initial value chosen for f j will have negligible effect on the forecast as the number of data points increases. However, the choice of a is very important and will affect the forecast accuracy. 

---

The moving average is a process similar to exponential smoothing. The exponential smoothing method (see also Section 6.4.2) has exponentially decreasing coefficients of the recent values. In a MA model the single coefficients b1 b2,. .., b were calculated by minimization of the sum of squared errors. 

Holt s two-parameter linear-exponential smoothing method is used to smooth trend and slope directly by using different smoothing constants for each. Using two constants gives more flexibility in selecting rates at which trend and slope can be followed. 

ES Agent, finally, determines forecasts according to the simple exponential smoothing method, which estimates the demand in any period as the weighted average of the last period demand and the forecast of demand in that period. It can be expressed... 

The predictive results of cubic exponential smoothing method... 

Liu Yang Ma Fenghai. 2007. Application of cubic exponential smoothing method to city underground deformation prediction. Technology Economy in Areas of Communicatitms (5) 62-70. 

Zhu Qingming Hao Zhang. 2012. Study on the application of cubic exponential smoothing method in coal mine accidents forecasting. Joural of Safety Science and Technology (4) 105-107. 

Zhang Weihong Li liu. 2008. Application of exponential smoothing method in the sales budget. China Management Information (2) 85-86. 

Step 4 Apply the exponential smoothing method on the deseasonalized demand to get deseasonalized forecast as shown in Table 2.3. For example, deseasonalized forecast for Quarter 1 (2011) is given by... 

Regular exponential smoothing model estimates the constant level (L) to forecast future demands. Holt s model improves it by estimating both the level (L) and trend factor (T). It adjusts both the level and trend factor using exponential smoothing. Holt s method is also known as double exponential smoothing or trend adjusted exponential smoothing method. 

Under the exponential smoothing method, the estimate of the level for (f + 1)... 

In Example 2.3, we then used the exponential smoothing method with a = 0.2 to determine the deseasonalized forecasts. Instead, we will apply Holt s method to incorporate trend also. 

Consider the actual demand, its forecast, and the errors for five periods given in Table 2.9. The forecasts are obtained using exponential smoothing method with a = 0.1. Compute MAD, MSE, STD, Bias, and MAPE for measuring the accuracy of the forecasting method. 

Interested readers can refer to the original paper by Croston (1972). Willemain et al. (1994) did a comparative evaluation of Croston s method in forecasting intermittent demand in manufacturing using industrial data. They concluded that Croston s method was superior to the exponential smoothing method and was robust even under situations when Croston s model assumptions were violated. 

What are the advantages and disadvantages of the Exponential Smoothing method ... 

SIMPLE EXPONENTIAL SMOOTHING The simple exponential smoothing method is appropriate when demand has no observable trend or seasonality. In this case. 

---