CHAPTER 10

JUDGMENTAL FORECASTING AND FORECAST ADJUSTMENTS

ANSWERS TO ODD-NUMBERED PROBLEMS AND CASES
1. The Delphi method can be used in any forecasting situation where there is little or no
historical data and there is expert opinion (experience) available. Two examples might
be:
First year sales for a new product
Full capacity employment at a new plant
Potential difficulties associated with the Delphi method include:
ssembling the !right" group of experts.
#vercoming individual biases or agendas
$ot being able to arrange timely feedbac%

CASE 10-1: GOLDEN GARDENS RESTAURANT
&. ' (. )ue and *ill have tac%led a very tough business pro+ect: designing a restaurant
that will succeed. ,estaurants seem to come and go on a regular basis so their
planning efforts prior to opening are important.
They have already tried focus groups and have some ideas to add to their own.
)ince they have a number of -expert- friends. some way must be found to use this
expertise. The Delphi method suggests itself as a way to utili/e their friends0
%nowledge. written description of the pro+ect along with the 1uestion of proper
motif could be supplied to each of their friends. along with a re1uest to design the
restaurant. These descriptions would then be mailed bac% to each participant with
a re1uest to re2design the business based on all the written replies. This process
could be continued until changes are no longer generated.
n optional step would then be to bring the participants together for a discussion.
This expert focus group could argue their cases and respond to )ue and *ill0s
ob+ections or insights. t the end of this process )ue and *ill would probably have
a better idea of how a successful restaurant would loo% and could begin their pro+ect
with more confidence. lso. financial bac%ers would probably be more enthusiastic
after reviewing the extensive planning that )ue and *ill have underta%en prior to
opening their business.
CASE 10-2: ALOMEGA FOOD STORES
&34
1. The na5ve forecasting model is not very accurate. The MSE e1uals
6.786.983.(:4.
3. ;f the na5ve approach had been more accurate. combining methods would have
been worth a try.
CASE 10-3: THE LYDIA E. PINKHAM MEDICINE COMPANY REVISITED
&. These articles are more abundant than many reali/e. <ore -popular" +ournals.
particularly financial mar%ets titles such as Technical nalysis of )toc%s '
=ommodities. Financial nalysts >ournal. and Futures present several articles.
;n addition. the proceedings from the neural networ% conferences (published by
;???) will usually have some business applications. Finally. this approach is
beginning to appear in more scholarly +ournals such as <anagement )cience and
Decision )ciences.
4. <odel specification is as much an art as it is a science. for example. loo% at =ase
@28 where the choice between ,;<(&.&.9) and ,(() models is not clearcut. $eural
networ%s. however. do not re1uire the analyst to specify the form of the
model 22 they have been called -model free- function approximators (see *art
Aos%o. Neural Networks and Fuzzy Systems: A Dynamical Systems Approach
to Machine Intelligence. Prentice2Ball. &@@(. for example).
CHAPTER 11
&38
MANAGING THE FORECASTING PROCESS
ANSWERS TO PROBLEMS AND CASES
1. a. #ne response: Forecasts may not be right. but they improve the odds of being
close to right. <ore importantly. if there are no agreed upon set of forecasts to
drive planning. then different groups may develop own procedures to guide
planning with potential chaos as the result.
b. #ne response: nalogyC;f you thin% education is expensive. try ignorance.
Baving a good set of forecasts is li%e wal%ing while loo%ing ahead instead of at
your shoes. Planning without forecasts will lead to inefficient operations. sub
optimal returns on investment. poor customer service. and so forth.

c. #ne response: Dood forecasts re1uire not only good 1uantitative s%ills. they also
re1uire an in2depth understanding of the business or. more generally. the
forecasting environment and. ultimately. good communication s%ills to sell
forecasts to management.
CASE 11-1: BOUNDARY ELECTRONICS
&. This case invites students to thin% about how to use some of the forecasting techni1ues
discussed in =hapter &&. Duy Preston is trying to get his managers to thin% about the
long2range position of the company. as opposed to the short range thin%ing that most
managers are involved in on a daily basis. The case might generate a class discussion
about the tendency of managers to shorten their planning hori/ons too much in the
daily press of business.
Duy has as%ed his managers to write scenarios for the future: a worst case. a status 1uo.
and a most li%ely scenario. Bis next tas% might be to discuss each of these three
possibilities. and to discuss any differences of opinion that might emerge. second
round of written scenarios by each participant could then follow this.
4. There are two possible benefits from Duy0s retreat. First. he may gain valuable insights
into the company0s future to use in his own long range thin%ing. )econd. and
probably more important. his managers may come away with an increased
awareness of the importance of expanding their planning hori/ons. ;f this is true.
the company will probably be in a better position to face the future.
CASE 11-2: BUSBY ASSOCIATES
&. )ince BoltEs linear smoothing incorporates simple exponential smoothing as a special
case (F G 9). would expect BoltEs procedure to fit and forecast better here.
Therefore. there is no reason to consider a combination of forecasts. =ombining
forecasts is best considered when the sets of forecasts are produced by different
&3:
procedures.

4. fter the results for a few additional 1uarters (say 8) become available. the analysis
can be re2done to see if the current model is still viable. <odel parameters can be
re2estimated after each new observation if appropriate computer software is available.
CASE 11-3: CONSUMER CREDIT COUNSELING
)tudents should summari/e the results of the analyses of these data in the cases at the
ends of chapters 8 (smoothing). : (decomposition). 7 (simple linear regression). 6 (regression
with time series data) and @ (*ox2>en%ins methods). Fits. residual analyses. and forecasts can be
compared. ,egardless of the method. there is a fair amount of unexplained variation in the
number of new clients. This may be a situation where combining forecasts ma%es sense.
CASE 11-: MR. TU!
He collected the data from the <r. Tux rental shop so that real data could be used at the
end of each chapter instead of contrived data. He didn0t %now what would happen when we tried
to forecast this variable. but we thin% it turned out well because no one method was superior.
The case in =hapter && summari/es the different ways >ohn used to forecast his monthly
sales. and as%s students to comment on his efforts. He thin% a %ey point is that a lot of real data
sets do not lend themselves to accurate forecasting. and that continually trying different methods is
re1uired. For the <r. Tux data. there are fairly simple seasonal models (see the cases in =hapters
6 and @) that represent the data well and provide reasonable forecasts.
Hhat advice should we give to >ohn <osby for the futureI )ome suggestions to offer
might include:
&. Jpdate the data set as future monthly values become available and re2run the
most promising analyses to see if the current forecasting model is still viable.
(. =onsider combining forecasts from two different methods.
4. Try to develop a useful relationship between monthly sales and regional
economic variables. Perhaps the area unemployment rate or an economic activity
index would correlate well with >ohn0s sales. Perhaps some demographic
variables would correlate well. ;f several variables were collected over the
months of >ohn0s sales data. a good regression e1uation might result.
This would allow >ohn to understand how is sales are tied to the local
environment.
CASE 11-5: ALOMEGA FOOD STORES
&. >ulie has to choose between two different methods of forecasting her companyEs
monthly sales. )tudents should review the results of these two efforts and decide
which offers the better choice. He find that class presentations by student teams
are valuable as they move the analysis beyond the computer results to simulate
&37
implementing these results in a !real" situation.
4. #ther forecasting methods are certainly possible in this case. n assignment
beyond a consideration of choosing between decomposition and multiple
regression would be to find a superior forecasting method using any available
software. gain. the 1ualitative considerations should be considered. including the
necessity of balancing the complexity and accuracy of a forecasting method with its
acceptance and use by the management team.
CASE 11-6: SOUTHWEST MEDICAL CENTER
)tudents should summari/e the results of <aryEs forecasting efforts describing the fits.
residual analyses and forecasts. <oreover. they should point out the apparent difficulty in
finding an ade1uate model for <aryEs total visits series. ;f <aryEs data is accurateCthere is no
reason for the apparent inconsistency in her time seriesCthen it would probably be wise to
collect another year or so of data and attempt to model the entire data set or. perhaps. +ust the
data following fiscal year 7. ;n the interim. she may have to settle for the forecasts from the
best ,;< model developed in =ase @2&9.
&33