Forecasting Methods: Moving Average, Exponential Smoothing, Linear Trend and Regression Forecast Error = Demand ( Actual) Forecast MAD (mean absolute deviation) = Average of absolute (positive) forecast errors, on average how much we are off MSE (mean squared error) = Average of squared errors MAPE (mean absolute \% error) = average of (absolute error/demand) You have been given the demand data for the past 10 weeks for swim rings for children. You decide to run three different types of forecasting methods on the data to see which gives you best forecast. Use 2 week moving average; weighted moving average with weights 0.7 and 0.3; exponential smoothing with alpha =.3, and Regression equation. Then calculate MAD, MSE, and MAPE for MA, ES and Regression Forecasts. Which forecasts are more accurate? Explain. What will be the forecasts for Week 22? 1. MOVING AVERAGE, MMAn=Di where Di= demand in period i,n : number of periods in the moving average a. Calculate 2 weeks MA forecasts MSE: MAPE