CHAPTER EIGHTEEN FORECASTING

MULTIPLE CHOICE QUESTIONS In the following multiple choice questions, circle the correct answer. 1. The time series component which reflects a regular, multi-year pattern of being above and below the trend line is a. a trend b. seasonal c. cyclical d. irregular The time series component that reflects variability during a single year is called a. a trend b. seasonal c. cyclical d. irregular The time series component that reflects variability due to natural disasters is called a. a trend b. seasonal c. cyclical d. irregular The time series component that reflects gradual variability over a long time period is called a. a trend b. seasonal c. cyclical d. irregular The trend component is easy to identify by using a. moving averages b. e"ponential smoothing c. regression analysis d. the #elphi approach The forecasting method that is appropriate when the time series has no significant trend, cyclical, or seasonal effect is a. moving averages b. mean squared error c. mean average deviation d. qualitative forecasting methods 1

2.

3.

!.

$.

%hapter &ighteen

'.

If data for a time series analysis is collected on an annual basis only, which component may be ignored( a. trend b. seasonal c. cyclical d. irregular *or the following time series, you are given the moving average forecast. Time Period 1 2 3 ! $ ' Time Series Value 23 1' 1' 2$ 11 23 1' Movin Avera e Fore!as"

).

1+ 2, 1) 2,

The mean squared error equals a. , b. $ c. 1 d. 1$ +. If the estimate of the trend component is 1!).2, the estimate of the seasonal component is + -, the estimate of the cyclical component is 1,!-, and the estimate of the irregular component is +)-, then the multiplicative model will produce a forecast of a. 1.!3 b. 1.!3c. 1!3.,2 d. 1!3,,2,,!32 .elow you are given the first four values of a time series. Time Period 1 2 3 Time Series Value 1) 2, 2! 1'

1,.

/sing a -period moving average, the forecasted value for period ! is a. 2.! b. 1' c. 2, d. 1,

*orecasting 11. .elow you are given the first two values of a time series. 0ou are also given the first two values of the e"ponential smoothing forecast. Time Period &"' 1 2 Time Series Value &("' 1) 22 E#$onen"ial Smoo"%in Fore!as" &F"' 1) 1)

If the smoothing constant equals .3, then the e"ponential smoothing forecast for time period three is a. 1) b. 1+.2 c. 2, d. , 12. The following linear trend e"pression was estimated using a time series with 1' time periods. Tt 1 12+.2 2 3.)t The trend pro3ection for time period 1) is a. $). b. 1+3.) c. 1+'.$ d. $.) E#%i)i" *+,* .elow you are given the first five values of a quarterly time series. The multiplicative model is appropriate and a four-quarter moving average will be used. (ear 1 Quar"er 1 2 3 1 Time Series Value (" 3$ 2 1$ 2,

2 13.

4efer to &"hibit 1)-1. 5n estimate of the trend component times the cyclical component 6T2%t7 for 8uarter 3 of 0ear 1, when a four-quarter moving average is used, is a. 2 b. 2! c. 2$ d. 2) 4efer to &"hibit 1)-1. 5n estimate of the seasonal-irregular component for 8uarter 3 of 0ear 1 is a. .$

1 .

%hapter &ighteen b. 1.!$2! c. !.333 d. 3, 1!. 0ou are given the following information on the seasonal-irregular component values for a quarterly time series9 Quar"er 1 2 3 Seasonal,Irre ular Com$onen" Values &S"I"' 1.23, 1.1!, 1.1$ .)$, .)+, .)3 .'', .'2, .'+ 1.2,, 1.13, 1.1'

The seasonal inde" for 8uarter 1 is a. .++' b. 1.1) c. d. 3 1$. .elow you are given some values of a time series consisting of 2$ time periods. Time Period 1 2 3 . . . 23 2 2! 2$ Time Series Value 3' ) !, $3

1,! 1,' 112 11

The estimated regression equation for these data is 0t 1 1$.23 2 .!20t-1 2 .3'0t-2 The forecasted value for time period 2' is a. !3.23 b. 1,+.! c. 11$.$! d. 11$.+! 1'. 5 group of observations measured at successive time intervals is :nown as a. a trend component b. a time series

*orecasting c. a forecast d. an additive time series model 1). 5 component of the time series model that results in the multi-period above-trend and below-trend behavior of a time series is a. a trend component b. a cyclical component c. a seasonal component d. an irregular component The model that assumes that the actual time series value is the product of its components is the a. forecast time series model b. multiplicative time series model c. additive time series model d. ;one of these alternatives is correct. 5 method of smoothing a time series that can be used to identify the combined trend<cyclical component is a. the moving average b. the percent of trend c. e"ponential smoothing d. the trend<cyclical inde" 5 method that uses a weighted average of past values for arriving at smoothed time series values is :nown as a. a smoothing average b. a moving average c. an e"ponential average d. an e"ponential smoothing In the linear trend equation, T 1 b, 2 b1t, b1 represents the a. trend value in period t b. intercept of the trend line c. slope of the trend line d. point in time In the linear trend equation, T 1 b, 2 b1t, b, represents the a. time b. slope of the trend line c. trend value in period 1 d. the 0 intercept

1+.

2,.

21.

22.

23.

2 .

5 parameter of the e"ponential smoothing model which provides the weight given to the most recent time series value in the calculation of the forecast value is :nown as the a. mean square error b. mean absolute deviation

%hapter &ighteen c. smoothing constant d. ;one of these alternatives is correct.

2!.

=ne measure of the accuracy of a forecasting model is a. the smoothing constant b. a deseasonali>ed time series c. the mean square error d. ;one of these alternatives is correct. 5 qualitative forecasting method that obtains forecasts through ?group consensus? is :nown as the a. 5utoregressive model b. #elphi approach c. mean absolute deviation d. ;one of these alternatives is correct.

2$.

E#%i)i" *+,%onsider the following time series. t 0i 2'. 1 2 ' 3 + 1,

4efer to &"hibit 1)-2. The slope of linear trend equation, b1, is a. 2.! b. 2., c. 1., d. 1.2! 4efer to &"hibit 1)-2. The intercept, b,, is a. 2.! b. 2., c. 1., d. 1.2! 4efer to &"hibit 1)-2. The forecast for period ! is a. 1,., b. 2.! c. 12.! d. .! 4efer to &"hibit 1)-2. The forecast for period 1, is a. 1,., b. 2!., c. 3,., d. 22.!

2).

2+.

3,.

E#%i)i" *+,. %onsider the following time series.

*orecasting

'

(ear &"' 1 2 3 ! 31.

(i ' ! 2 1

4efer to &"hibit 1)-3. The slope of linear trend equation, b1, is a. -1.! b. 21.! c. ).3 d. -).3 4efer to &"hibit 1)-3. The intercept, b,, is a. -1.! b. 21.! c. ).3 d. -).3 4efer to &"hibit 1)-3. In which time period does the value of 0i reach >ero( a. ,.,,, b. ,.1)1 c. !.!3 d. .21 4efer to &"hibit 1)-3. The forecast for period 1, is a. $.' b. -$.' c. 23.3 d. 1!

32.

33.

3 .

%hapter &ighteen

PRO/LEMS 1. The sales records of a company over a period of seven years are shown below. (ear &"' 1 2 3 ! $ ' Sales &In Millions o0 1ollars' 12 1$ 1' 1+ 1) 21 22

a. #evelop a linear trend e"pression for the above time series. b. *orecast sales for period 1,. 2. @tudent enrollment at a university over the past si" years is given below. (ear &"' 1 2 3 ! $ Enrollmen" &In *2333s' $.3, '.', ).,, ).2, ).), ).,,

a. #evelop a linear trend e"pression for the above time series. b. *orecast enrollment for year 1,. 3. The following time series shows the sales of a clothing store over a 1,-wee: period. 4ee5 1 2 3 ! $ ' ) + 1, Sales &6*2333s' 1! 1$ 1+ 1) 1+ 2, 1+ 22 1! 21

a. %ompute a -wee: moving average for the above time series.

*orecasting b. %ompute the mean square error 6A@&7 for the -wee: moving average forecast. c. /se 1 ,.3 to compute the e"ponential smoothing values for the time series. d. *orecast sales for wee: 11. . The following time series shows the number of units of a particular product sold over the past si" months. Mon"% 1 3 3 ! $ a. b. c. d. !. Uni"s Sold &T%ousands' ) ! 12 1,

%ompute a 3-month moving average 6centered7 for the above time series. %ompute the mean square error 6A@&7 for the 3-month moving average. /se 1 ,.2 to compute the e"ponential smoothing values for the time series. *orecast the sales volume for month '.

The sales volumes of %AA, Inc., a computer firm, for the past ) years is given below. (ear &"' 1 2 3 ! $ ' ) Sales &In Millions o0 1ollars' 2 3 ! $ ) + +

a. #evelop a linear trend e"pression for the above time series. b. *orecast sales for period +. $. The sales records of a ma3or auto manufacturer over the past ten years are shown below. (ear &"' 1 2 3 Num)er o0 Cars Sold &In "%ousands o0 Uni"s' 1+! 2,, 2!,

1,

%hapter &ighteen 2', 32, 3), , $, !,, !,,

! $ ' ) + 1,

#evelop a linear trend e"pression and pro3ect the sales 6the number of cars sold7 for time period t 1 11. '. The following data show the quarterly sales of 5ma>ing Braphics, Inc. for the years $ through ). (ear $ Quar"er 1 2 3 1 2 3 1 2 3 Sales 2.! 1.! 2. 1.$ 2., 1. 1.' 1.+ 2.! 2., 2. 2.1

'

a. %ompute the four-quarter moving average values for the above time series. b. %ompute the seasonal factors for the four quarters. c. /se the seasonal factors developed in Cart b to ad3ust the forecast for the effect of season for year $. ). Dohn has collected the following information on the amount of tips he has collected from par:ing cars the last seven nights. 1a7 1 2 3 ! $ ' Ti$s 1) 22 1' 1) 2) 2, 12

a. %ompute the 3-day moving averages for the time series.

*orecasting b. %ompute the mean square error for the forecasts. c. %ompute the mean absolute deviation for the forecasts. d. *orecast DohnEs tips for day '. +.

11

The following information has been collected on the sales of greeting cards for the past $ wee:s. 4ee5 1 2 3 ! $ Sales 1,! +, +! 11, 1,! 1,,

a. Croduce e"ponential smoothing forecasts for the series using a smoothing constant of .2. b. %ompute the mean square error for the forecasts produced with a smoothing constant of .2. c. Fhat is the forecast of sales for wee: '( d. Is a smoothing constant of .2 or .3 better for the sales data( &"plain. 1,. %onsider the following annual series on the number of people assisted by a county human resources department. (ear 1 2 3 ! $ ' ) + 1, 11 Peo$le &in *33s' 22 2 2) 2 22 2 2, 2$ 2 2) 2$

a. Crepare 3-year moving average values to be used as forecasts for periods through 11. %alculate the mean squared error 6A@&7 measure of forecast accuracy for periods through 11. b. /se a smoothing constant of . to compute e"ponential smoothing values to be used as forecasts for periods 2 through 11. %alculate the A@&. c. %ompare the results in Carts a and b. 11. The temperature in %hicago has been recorded for the past seven days. 0ou are given the information below.

12

%hapter &ighteen 1a7 1 2 3 ! $ ' Tem$era"ure )2 ), ) )3 ), '+ )2

a. Croduce e"ponential smoothing forecasts for the series using a smoothing constant of .2. b. %ompute the mean square error for the forecasts produced with a smoothing constant of .2. c. Fhat is the forecasted temperature for day )( d. Is a smoothing constant of .2 or .3 better for the temperature data( &"plain. 12. The yearly series below e"hibits a long-term trend. /se the appropriate forecasting technique to produce forecasts for years 11 and 12. (ear 1 2 3 ! $ ' ) + 1, 13. Time Series Value 12, 132 1 ) 1!2 1$, 1'! 1)2 1+, 1+! 2,!

The following time series gives the number of units sold during ! years at a boat dealership. (ear 1 Quar"er 1 2 3 1 2 3 1 2 3 1 Num)er o0 Uni"s 3,, 2 , 2 , 2+, 3!, 3,, 2), 32, 1, ,, 3+, 1, +,

*orecasting 2 3 ! 1 2 3 !, , !1, ! , !3, !2, ! ,

13

a. b. c. d. e. f. g. 1 .

*ind the four-quarter centered moving averages. Clot the series and the moving averages on a graph. %ompute the seasonal-irregular component. %ompute the seasonal factors for all four quarters. %ompute the deseasonali>ed time series for sales. %alculate the linear trend from the deseasonali>ed sales. *orecast the number of units sold in each quarter of year $.

.elow you are given information on DohnEs income for the past ' years. (ear 1 2 3 ! $ ' In!ome &In T%ousands' 1!., 1$.2 1'.1 1).1 1).) 1+.2 2,.!

a. /se regression analysis to obtain an e"pression for the linear trend component. b. *orecast DohnEs income for the ne"t ! years. 1!. 0ou are given the following information on the quarterly profits for 53a" %orporation. (ear 1 Quar"er 1 2 3 1 2 3 1 2 3 1 2 Quar"erl7 Pro0i"s (" 1!, 12, 1$, 1!, 1!, 13, 1), 1$, 1', 1 , 2,, 1), 2,, 1!,

%hapter &ighteen 3 a. b. c. d. 23, 2,,

*ind the four-quarter centered moving averages. %ompute the seasonal-irregular component. %ompute the seasonal factors for all four quarters. 4epresent the deseasonali>ed series.

1$.

.elow you are given information on crime statistics for Aiddletown. (ear 1 Quar"er 1 2 3 1 2 3 1 2 3 1 2 3 The seasonal factors for these data are Quar"er 1 2 3 Seasonal Fa!"or S" .!)+ 1.3!1 1.33! .'2$ Num)er o0 Crimes Commi""ed (" 1, 2, 2! ! 1, 3, 3! 2! 2, , 3! 1! 2, !, ! 3!

a. #eseasonali>e the series. b. =btain an estimate of the linear trend for this series. c. /se the seasonal and trend components to forecast the number of crimes for each quarter of 0ear !. 1'. .elow you are given the seasonal factors and the estimated trend equation for a time series. These values were computed on the basis of ! years of quarterly data. Quar"er Seasonal Fa!"or S"

*orecasting 1 2 3 T 1 12$.23 - 1.$t Croduce forecasts for all four quarters of year $ by using the seasonal and trend components. 1). The following data show the quarterly sales of a ma3or auto manufacturer 6introduced in e"ercise 7 for the years ) through 1,. (ear ) Quar"er 1 2 3 1 2 3 1 2 3 Sales 1$, 1), 1+, 1', 2,, 21, 2$, 23, 21, 2 , 2+, 2$, 1.2 .+ .) 1.1

1!

1,

a. %ompute the four-quarter moving average values for the above time series. b. %ompute the seasonal factors for the four quarters. c. /se the seasonal factors developed in Cart b to ad3ust the forecast for the effect of season for year +.