1. 1 1 Slide © 2009 South-Western, a part of Cengage Learning Chapter 6 Forecasting n Quantitative Approaches to Forecasting n Components of a Time Series n Measures of Forecast Accuracy n Smoothing Methods n Trend Projection n Trend and Seasonal Components n Regression Analysis n Qualitative Approaches

2. 2 2 Slide © 2009 South-Western, a part of Cengage Learning Quantitative Approaches to Forecasting n A forecast is a prediction of what will happen in the future. n A time series is a set of observations measured at successive points in, or over successive periods of, time. n Quantitative forecasting approaches are based on an analysis of historical data concerning one or more time series. n If the data used are limited to past values of the series we are trying to forecast, the procedure is called a time series method. n If the data used involve other time series that are believed to be related to the time series being forecasted, the procedure is called a causal method.

3. 3 3 Slide © 2009 South-Western, a part of Cengage Learning Time Series Methods n The objective of time series methods is to discover a pattern in the historical data and then extrapolate this pattern into the future. n The forecast is based solely on past values of the variable that we are trying to forecast and/or on past forecast errors. n Three time series methods are: smoothing smoothing trend projection trend projection trend projection adjusted for seasonal influence trend projection adjusted for seasonal influence

4. 4 4 Slide © 2009 South-Western, a part of Cengage Learning Components of a Time Series n The trend component accounts for the gradual shifting of the time series over a long period of time. n Any regular pattern of sequences of values above and below the trend line is attributable to the cyclical component of the series.

5. 5 5 Slide © 2009 South-Western, a part of Cengage Learning Components of a Time Series n The seasonal component of the series accounts for regular patterns of variability within certain time periods, such as over a year. n The irregular component of the series is caused by short-term, unanticipated and non-recurring factors that affect the values of the time series. One cannot attempt to predict its impact on the time series in advance.

6. 6 6 Slide © 2009 South-Western, a part of Cengage Learning Smoothing Methods n In cases in which the time series is fairly stable and has no significant trend, seasonal, or cyclical effects, one can use smoothing methods to average out the irregular components of the time series. n Three common smoothing methods are: Moving averages Moving averages Weighted moving averages Weighted moving averages Exponential smoothing Exponential smoothing

7. 7 7 Slide © 2009 South-Western, a part of Cengage Learning Smoothing Methods n Moving Average Method The moving average method consists of computing an average of the most recent n data values in the series and using this average for forecasting the value of the time series for the next period. The term moving indicates that, as a new observation becomes available for the time series, it replaces the oldest observation in the equation, and a new average is computed.

8. 8 8 Slide © 2009 South-Western, a part of Cengage Learning The data below show the number of gallons of gasoline (in 1,000s) sold by a gasoline distributor in Bennington, Vermont, over the past 12 weeks. The time series appears to be stable over time, so smoothing methods are applicable. Example: Gasoline Sales Week Sales Week Sales Week Sales Week Sales 1 17 7 20 1 17 7 20 2 21 8 18 2 21 8 18 3 19 9 22 3 19 9 22 4 23 10 20 4 23 10 20 5 18 11 15 5 18 11 15 6 16 12 22 6 16 12 22

9. 9 9 Slide © 2009 South-Western, a part of Cengage Learning n Moving Average To forecast gasoline sales for week 13 (in 1,000s) using a 3-week moving average, we need to compute the average of sales for weeks 10, 11, and 12. The week 13 forecast (F 13 ) is: Example: Gasoline Sales

10. 10 Slide © 2009 South-Western, a part of Cengage Learning Smoothing Methods n Weighted Moving Average Method In the weighted moving average method for computing the average of the most recent n periods, the more recent observations are typically given more weight than older observations. For convenience, the weights usually sum to 1.

11. 11 Slide © 2009 South-Western, a part of Cengage Learning n Three-Period Weighted Moving Average To forecast gasoline sales (in 1,000s) for week 13 using a 3-week weighted moving average, we will compute the weighted average of sales for weeks 10, 11, and 12 using the weights of 1/6, 2/6, and 3/6. The week 13 forecast (F 13 ) is: To forecast gasoline sales (in 1,000s) for week 13 using a 3-week weighted moving average, we will compute the weighted average of sales for weeks 10, 11, and 12 using the weights of 1/6, 2/6, and 3/6. The week 13 forecast (F 13 ) is: F 13 = 1/6(20) + 2/6(15) + 3/6(22) F 13 = 1/6(20) + 2/6(15) + 3/6(22) = 3.333 + 5.000 + 11.000 = 3.333 + 5.000 + 11.000 = 19.333 = 19.333 Example: Gasoline Sales

12. 12 Slide © 2009 South-Western, a part of Cengage Learning Smoothing Methods n Exponential Smoothing Using exponential smoothing, the forecast for the next period is equal to the forecast for the current period plus a proportion (  ) of the forecast error in the current period. Using exponential smoothing, the forecast for the next period is equal to the forecast for the current period plus a proportion (  ) of the forecast error in the current period. The forecast is calculated by: The forecast is calculated by:  [the actual value for the current period] +  [the actual value for the current period] + (1-  )[the forecasted value for the current period], (1-  )[the forecasted value for the current period], where the smoothing constant, , is a number between 0 and 1.

13. 13 Slide © 2009 South-Western, a part of Cengage Learning n Exponential Smoothing F t +1 =  Y t + (1 –  ) F t F t +1 =  Y t + (1 –  ) F twhere: F t+1 = forecast of the time series for period t +1 F t+1 = forecast of the time series for period t +1 Y t = actual value of the time series in period t Y t = actual value of the time series in period t F t = forecast of the time series for period t F t = forecast of the time series for period t  = smoothing constant (0 <  < 1)  = smoothing constant (0 <  < 1) (To start the calculations, we let F 1 equal the (To start the calculations, we let F 1 equal the actual value of the time series in period 1.) Smoothing Methods

14. 14 Slide © 2009 South-Western, a part of Cengage Learning Exponential Smoothing (  =.2, 1 -  =.8) Exponential Smoothing (  =.2, 1 -  =.8) F 1 = 17 F 2 =.2 Y 1 +.8 F 1 =.2(17) +.8(17) = 17.00 F 3 =.2 Y 2 +.8 F 2 =.2(21) +.8(17) = 17.80 F 4 =.2 Y 3 +.8 F 3 =.2(19) +.8(17.80) = 18.04 F 5 =.2 Y 4 +.8 F 4 =.2(23) +.8(18.04) = 19.03 F 6 =.2 Y 5 +.8 F 5 =.2(18) +.8(19.03) = 18.83 F 7 =.2 Y 6 +.8 F 6 =.2(16) +.8(18.83) = 18.26 Example: Gasoline Sales

15. 15 Slide © 2009 South-Western, a part of Cengage Learning Example: Gasoline Sales Exponential Smoothing (  =.2, 1 -  =.8) Exponential Smoothing (  =.2, 1 -  =.8) F 8 =.2 Y 7 +.8 F 7 =.2(20) +.8(18.26) = 18.61 F 9 =.2 Y 8 +.8 F 8 =.2(18) +.8(18.61) = 18.49 F 10 =.2 Y 9 +.8 F 9 =.2(22) +.8(18.49) = 19.19 F 11 =.2 Y 10 +.8 F 10 =.2(20) +.8(19.19) = 19.35 F 12 =.2 Y 11 +.8 F 11 =.2(15) +.8(19.35) = 18.48 F 13 =.2 Y 12 +.8 F 12 =.2(22) +.8(18.48) = 19.18

16. 16 Slide © 2009 South-Western, a part of Cengage Learning Mean Squared Error (MSE) n An important consideration in selecting a forecasting method is the accuracy of the forecast. Clearly, we want forecast errors to be small. n The mean squared error (MSE) is an often-used measure of the accuracy of a forecasting method. n The average of the squared forecast errors for the historical data is calculated. n The forecasting method or parameter(s) which minimize this mean squared error is then selected.

17. 17 Slide © 2009 South-Western, a part of Cengage Learning Example: Gasoline Sales n Mean Squared Error

18. 18 Slide © 2009 South-Western, a part of Cengage Learning Trend Projection n If a time series exhibits a linear trend, the method of least squares may be used to determine a trend line (projection) for future forecasts. n Least squares, also used in regression analysis, determines the unique trend line forecast which minimizes the mean square error between the trend line forecasts and the actual observed values for the time series. n The independent variable is the time period and the dependent variable is the actual observed value in the time series.

19. 19 Slide © 2009 South-Western, a part of Cengage Learning Trend Projection n Using the method of least squares, the formula for the trend projection is: T t = b 0 + b 1 t. where: T t = trend value in period t where: T t = trend value in period t b 0 = intercept of the trend line b 0 = intercept of the trend line b 1 = slope of the trend line b 1 = slope of the trend line where: Y t = actual value of the time series in period t where: Y t = actual value of the time series in period t = average value of the time series = average value of the time series = average value of t = average value of t n = number of periods n = number of periods

20. 20 Slide © 2009 South-Western, a part of Cengage Learning Consider the time series for bicycle sales (in 1,000s) of a particular manufacturer over the past 10 years, as shown below. Although the data show some up-and- down movement over the past 10 years, the time series for the number of bicycles sold seems to have an overall increasing or upward trend. Example: Bicycle Sales Year Sales Year Sales Year Sales Year Sales 1 21.6 6 27.5 1 21.6 6 27.5 2 22.9 7 31.5 2 22.9 7 31.5 3 25.5 8 29.7 3 25.5 8 29.7 4 21.9 9 28.6 4 21.9 9 28.6 5 23.9 10 31.4 5 23.9 10 31.4

21. 21 Slide © 2009 South-Western, a part of Cengage Learning Example: Bicycle Sales n Trend Projection t Y t tY t t 2 t Y t tY t t 2 1 21.6 21.6 1 1 21.6 21.6 1 2 22.9 45.8 4 2 22.9 45.8 4 3 25.5 76.5 9 3 25.5 76.5 9 4 21.9 87.6 16 4 21.9 87.6 16 5 23.9 119.5 25 5 23.9 119.5 25 6 27.5 165.0 36 6 27.5 165.0 36 7 31.5 220.5 49 7 31.5 220.5 49 8 29.7 237.6 64 8 29.7 237.6 64 9 28.6 257.4 81 9 28.6 257.4 81 10 31.4 314.0 100 10 31.4 314.0 100 55 264.5 1545.5 385 55 264.5 1545.5 385

22. 22 Slide © 2009 South-Western, a part of Cengage Learning n Trend Projection (continued) = 55/10 = 5.5 = 264.5/10 = 26.45 = 55/10 = 5.5 = 264.5/10 = 26.45 n  tY t -  t  Y t (10)(1545.5) - (55)(264.5) n  tY t -  t  Y t (10)(1545.5) - (55)(264.5) b 1 = = = 1.1 b 1 = = = 1.1 n  t 2 - (  t ) 2 (10)(385) - (55) 2 n  t 2 - (  t ) 2 (10)(385) - (55) 2 = 26.45 - 1.1(5.5) = 20.4 = 26.45 - 1.1(5.5) = 20.4 T t = 20.4 + 1.1( t ) T t = 20.4 + 1.1( t ) T 11 = 20.4 + 1.1(11) = 32.5 (or 32,500 bicycles) Example: Bicycle Sales

23. 23 Slide © 2009 South-Western, a part of Cengage Learning Forecasting with Trend and Seasonal Components n Steps of Multiplicative Time Series Model 1. Calculate the centered moving averages (CMAs). 2. Center the CMAs on integer-valued periods. 3. Determine the seasonal and irregular factors ( S t I t ). 4. Determine the average seasonal factors. 5. Scale the seasonal factors ( S t ). 6. Determine the deseasonalized data. 7. Determine a trend line of the deseasonalized data. 8. Determine the deseasonalized predictions. 9. Take into account the seasonality.

24. 24 Slide © 2009 South-Western, a part of Cengage Learning Example: Television Sales The data below show quarterly television sales (in thousands of units) for a particular manufacturer over the past four years. Quarter Year 1 2 3 4 Year 1 2 3 4 1 4.8 4.1 6.0 6.5 1 4.8 4.1 6.0 6.5 2 5.8 5.2 6.8 7.4 2 5.8 5.2 6.8 7.4 3 6.0 5.6 7.5 7.8 3 6.0 5.6 7.5 7.8 4 6.3 5.9 8.0 8.4 4 6.3 5.9 8.0 8.4

25. 25 Slide © 2009 South-Western, a part of Cengage Learning n 1. Calculate the centered moving averages. There are four distinct seasons in each year. Hence, take a four-period centered moving average to eliminate seasonal and irregular factors. For example: 1 st CMA = (4.8 + 4.1 + 6.0 + 6.5)/4 = 5.35 1 st CMA = (4.8 + 4.1 + 6.0 + 6.5)/4 = 5.35 2 nd CMA = (4.1 + 6.0 + 6.5 + 5.8)/4 = 5.60 2 nd CMA = (4.1 + 6.0 + 6.5 + 5.8)/4 = 5.60etc. Example: Television Sales

26. 26 Slide © 2009 South-Western, a part of Cengage Learning Example: Television Sales n 2. Center the CMAs on integer-valued periods. The first centered moving average computed in step 1 (5.35) will be centered on quarter 2.5 of year 1. The second centered moving average computed (5.60) will be centered on quarter 3.5 of year 1. Note that the moving averages from step 1 do not center themselves on integer-valued periods because n (the number of seasons) is an even number. Note that the moving averages from step 1 do not center themselves on integer-valued periods because n (the number of seasons) is an even number. In this case, 2-period CMAs are computed for the original 4-period CMAs. These 2-CMAs will center themselves on integer-valued periods.

27. 27 Slide © 2009 South-Western, a part of Cengage Learning Example: Television Sales n 2. Center the CMAs on integer-valued periods.

28. 28 Slide © 2009 South-Western, a part of Cengage Learning Example: Television Sales n 3. Determine the seasonal & irregular factors ( S t I t ). Isolate the trend and cyclical components. For each period t, this is given by: S t I t = Y t /(Moving Average for period t ) S t I t = Y t /(Moving Average for period t )

29. 29 Slide © 2009 South-Western, a part of Cengage Learning Example: Television Sales n 3. Determine the seasonal & irregular factors ( S t I t ).

30. 30 Slide © 2009 South-Western, a part of Cengage Learning Example: Television Sales n 4. Determine the average seasonal factors. Averaging all S t I t values corresponding to that season: Season 1: (0.971 + 0.918 + 0.908) /3 = 0.9323 Season 1: (0.971 + 0.918 + 0.908) /3 = 0.9323 Season 2: (0.840 + 0.839 + 0.834) /3 = 0.8377 Season 2: (0.840 + 0.839 + 0.834) /3 = 0.8377 Season 3: (1.096 + 1.075 + 1.109) /3 = 1.0933 Season 3: (1.096 + 1.075 + 1.109) /3 = 1.0933 Season 3: (1.133 + 1.156 + 1.141) /3 = 1.1433 Season 3: (1.133 + 1.156 + 1.141) /3 = 1.1433

31. 31 Slide © 2009 South-Western, a part of Cengage Learning Example: Television Sales n 5. Scale the seasonal factors ( S t ). Average the seasonal factors = (0.9323 + 0.8377 Average the seasonal factors = (0.9323 + 0.8377 + 1.0933 + 1.1433)/4 = 1.00165. Then, divide each seasonal factor by the average of the seasonal factors. + 1.0933 + 1.1433)/4 = 1.00165. Then, divide each seasonal factor by the average of the seasonal factors. Season 1: 0.9323/1.00165 = 0.9308 Season 1: 0.9323/1.00165 = 0.9308 Season 2: 0.8377/1.00165 = 0.8363 Season 2: 0.8377/1.00165 = 0.8363 Season 3: 1.0933/1.00165 = 1.0915 Season 4: 1.1433/1.00165 = 1.1414 Season 4: 1.1433/1.00165 = 1.1414 Total = 4.0000 Total = 4.0000

32. 32 Slide © 2009 South-Western, a part of Cengage Learning Example: Television Sales n 6. Determine the deseasonalized data. Divide the data point values, Y t, by S t.

33. 33 Slide © 2009 South-Western, a part of Cengage Learning Example: Television Sales n 6. Determine the deseasonalized data.

34. 34 Slide © 2009 South-Western, a part of Cengage Learning Example: Television Sales n 7. Determine a trend line of the deseasonalized data. Using the least squares method for t = 1, 2,..., 16, gives: Using the least squares method for t = 1, 2,..., 16, gives: T t = 5.101 + 0.148( t ) T t = 5.101 + 0.148( t )

35. 35 Slide © 2009 South-Western, a part of Cengage Learning Example: Television Sales n 8. Determine the deseasonalized (trend) predictions. Substitute t = 17, 18, 19, and 20 into the least squares equation: Year 5, Quarter 1: T 17 = 5.101 + 0.148(17) = 7.617 Year 5, Quarter 1: T 17 = 5.101 + 0.148(17) = 7.617 Year 5, Quarter 2: T 18 = 5.101 + 0.148(18) = 7.765 Year 5, Quarter 2: T 18 = 5.101 + 0.148(18) = 7.765 Year 5, Quarter 3: T 19 = 5.101 + 0.148(19) = 7.913 Year 5, Quarter 3: T 19 = 5.101 + 0.148(19) = 7.913 Year 5, Quarter 4: T 20 = 5.101 + 0.148(20) = 8.061 Year 5, Quarter 4: T 20 = 5.101 + 0.148(20) = 8.061

36. 36 Slide © 2009 South-Western, a part of Cengage Learning Example: Television Sales n 9. Take into account the seasonality. Multiply each deseasonalized (trend) prediction by its seasonal factor to give the following forecasts for year 5: Year 5, Quarter 1: F 17 = 7.617(0.9308) = 7.090 Year 5, Quarter 1: F 17 = 7.617(0.9308) = 7.090 Year 5, Quarter 2: F 18 = 7.765(0.8363) = 6.494 Year 5, Quarter 2: F 18 = 7.765(0.8363) = 6.494 Year 5, Quarter 3: F 19 = 7.913(1.0915) = 8.637 Year 5, Quarter 3: F 19 = 7.913(1.0915) = 8.637 Year 5, Quarter 4: F 20 = 8.061(1.1414) = 9.201 Year 5, Quarter 4: F 20 = 8.061(1.1414) = 9.201

37. 37 Slide © 2009 South-Western, a part of Cengage Learning Qualitative Approaches to Forecasting n Delphi Approach A panel of experts, each of whom is physically separated from the others and is anonymous, is asked to respond to a sequential series of questionnaires. A panel of experts, each of whom is physically separated from the others and is anonymous, is asked to respond to a sequential series of questionnaires. After each questionnaire, the responses are tabulated and the information and opinions of the entire group are made known to each of the other panel members so that they may revise their previous forecast response. After each questionnaire, the responses are tabulated and the information and opinions of the entire group are made known to each of the other panel members so that they may revise their previous forecast response. The process continues until some degree of consensus is achieved. The process continues until some degree of consensus is achieved.

38. 38 Slide © 2009 South-Western, a part of Cengage Learning Qualitative Approaches to Forecasting n Scenario Writing Scenario writing consists of developing a conceptual scenario of the future based on a well defined set of assumptions. Scenario writing consists of developing a conceptual scenario of the future based on a well defined set of assumptions. After several different scenarios have been developed, the decision maker determines which is most likely to occur in the future and makes decisions accordingly. After several different scenarios have been developed, the decision maker determines which is most likely to occur in the future and makes decisions accordingly.

39. 39 Slide © 2009 South-Western, a part of Cengage Learning Qualitative Approaches to Forecasting n Subjective or Interactive Approaches These techniques are often used by committees or panels seeking to develop new ideas or solve complex problems. These techniques are often used by committees or panels seeking to develop new ideas or solve complex problems. They often involve "brainstorming sessions". They often involve "brainstorming sessions". It is important in such sessions that any ideas or opinions be permitted to be presented without regard to its relevancy and without fear of criticism. It is important in such sessions that any ideas or opinions be permitted to be presented without regard to its relevancy and without fear of criticism.