T 18 07 Seasonally Adjusted Linear Trend Forecast

T 18 -07 Seasonally Adjusted Linear Trend Forecast Purpose Allows the analyst to create and analyze a "Seasonally Adjusted Linear Trend" forecast. The MAD and MSE for the forecast are calculated, and a various graphical displays are shown: seasonal relationship, historical demand forecast, as well as several averaging and de-seasonalized representations. Inputs Historical Time Series Seasonality Labels Outputs and Seasonally Forecast in the forecast method Limitations Adjusted Seasonal Relatives Graph showing Historical Time Series Graph showing Adjusted Seasonal Relatives Graph showing Historical Time Series Adjusted Linear Trend Various graphs showing steps 60 Time Series Observations 1

Time Series Techniques Seasonal Relatives – develops factors based on seasonality. These factors are used to adjust future forecasts Horizon: Intermediate range Method: Strength: Ability to determine a seasonality factors to adjust future forecasts. Weakness: Lot of effort when no seasonality exists. Good idea to look at data to determine if seasonality should be considered. 2

Forecast Accuracy Given that a forecast is rarely correct, the methodology you choose should be the one which provides the least error from the actual historical demand. Forecast error is defined as the difference between actual historical demand the forecast. 3

Forecast Error 4

Monitoring the Forecast There are two measures used to monitor the accuracy of a forecast. The Mean Absolute Deviation (MAD) and the Mean Squared Error (MSE). The MAD is the average of the absolute value of the forecast errors. The MSE is the average of the squared forecast errors. Note: The formula for the MSE shown above may vary slightly. Some textbooks divide the sum of the squared errors by n-1 rather than n. 5

Seasonally Adjusted Linear Trend Forecast Example A company has looked at its quarterly sales over the last three years, and believes that in addition to a linear trend a seasonal pattern is present. Determine the adjusted seasonal relatives and create a seasonally adjusted linear trend forecast. 6

Input the Seasonality Label and Time Series in the light green cells. The adjusted seasonal relatives and seasonal forecast are automatically calculated. 7

A graph showing the Time Series (Historical Demand) is automatically produced. 8

A graph showing the Time Series (Historical Demand), Moving Average, and Centered Moving Average is automatically produced. 9

A graph showing the Adjusted Seasonal Relatives is automatically produced. 10

A graph showing the Time Series (Historical Demand) and Deseasonalized Demand is automatically produced. 11

A graph showing the Time Series (Historical Demand), Deseasonalized Demand, and Deseasonlized Forecast is automatically produced. 12

A graph showing the Time Series (Historical Demand), Seasonally Adjusted Linear Trend Forecast is automatically produced. 13