Error in a forecasting approach is the difference between the actual data values and the forecasted values of that period. It helps to determine which forecasting method is suitable and provide better results than others. Error measures that can be applied to optimize the results of various forecasting methods include: mean absolute deviation MAD, mean squared error MSE, and mean absolute percentage error MAPE, and other measures like tracking signal for keeping/determining the values within a certain control limit. MAD = (sum of absolute forecast values/number of error periods) MSE (sum of the squares of error for each period) MAPE = (sum of absolute percentage values)/number of error periods. Tracking signal-recurring sum of forecast error/MAD These error measures provide different values for different methods applied to the forecasting problem. The smaller the error value, the more accurate and reliable will the forecast method. When the said methods of forecasting are used to obtain the forecast values, one can calculate all the error measures to know which method provides optimal results. It can be noted that, exponential smoothing forecast with a smoothing constant can provide better results in general. But, it can be subjective to the data. Simple moving average method is easier, but may not provide better accuracy, due to the lack of recency. Thus, making right selection of the method with optimized outcomes can be possible with error measures. Here you say " Simple moving average method is easier, but may not provide better accuracy, due to the lack of recency." What is an alternative to a simple moving average?