FORECAST WORKSHOP

1. The data is presented by quarters for the last two years. With this data, prepare,
through decomposition, a forecast for the following year.

Period Real Period Real

1 300 5 416
2 540 6 760
3 885 7 1191
4 580 8 760

2. Sunrise Baking Company sells donuts in a chain of food stores. Due to

forecast errors have led to excessive or insufficient production. The following
data is your demand for dozens of donuts in the last four weeks. The donuts are
they make for the next day; for example, the production of donuts on Sunday is for the
Monday sales, Monday's donut production is for Tuesday's sales, etc.
bakery closes on Saturdays, so the production on Friday must satisfy the
demand on Saturday and Sunday.

["4 weeks ago","3 weeks ago","2 weeks ago","last week"]

Monday 2200 2400 2300 2400
Tuesday 2000 2100 2200 2200
Wednesday 2300 2400 2300 2500
Thursday 1800 1900 1800 2000
Friday 1900 1800 2100 2000
Saturday
Sunday 2800 2700 3000 2900

Make a forecast for this week according to this scheme:

a. Daily, with a simple moving average of four weeks.
b. Daily, with a weighted moving average of 0.40, 0.30, 0.20, and 0.10 for the last
four weeks.
c. Sunrise also plans its purchases of ingredients for bread production.
If last week a demand for bread of 22,000 loaves was forecasted and only
21,000 were demanded, what should be the forecasted demand for Sunrise?
this week, with an exponential smoothing of = 0.10?
3. The demand for a product was forecasted using a specific model. The table shows
the forecasts and the corresponding demand that were presented. With DAM techniques.
and from the tracking signal, evaluate the accuracy of the forecasting model.
Real Forecasted
October 700 660
November 760 840
December 780 750
January 790 835
February 850 910
March 950 890
4. The demand for headphones for stereo devices and MP3 players for
treaders allowed Nina Industries to grow nearly 50% last year. The number of
Walkers continue to increase, so Nina hopes that the demand will also rise.
because, until now, no security laws have been enacted that prevent the
Runners use headphones. The demand for headphones last year was as follows:
Mes Demanda (unidades) My Demand (units)
January 4200 July 5300
February 4300 August 4900
March 4000 September 5400
April 4400 October 5700
May 5000 November 6300
June 4700 December 6000
a. With a least squares regression analysis, what would you estimate it to be?
demand for each month of the coming year? With a spreadsheet. Compare your
results obtained with the forecasting function of the sheet
calculation.
b. To have some assurance of meeting the demand, Nina decides to use three mistakes.
standard for safety. How many additional units should be retained to reach
this level of trust?
5. The historical demand for the product is:
Demand
January 12
February 11
March 15
April 12
may 16
June 15

a. With a three-month simple moving average, determine the forecast for July.
b. With a weighted moving average and weights of 0.60, 0.30, and 0.10, calculate the
July forecast.
c. Using simple exponential smoothing with alpha = 0.2 and a forecast for June of 13,
Calculate the forecast for July. Make any assumptions you want.
d. Calculate the exponential smoothing forecast with trend for these data, with
a .30, δ of .30, an initial trend forecast (T1) of 1 and a forecast
initial exponential smoothing (F1) of 14.
e. With a least squares regression analysis, what would you estimate it to be?
demand for each month of the next year?
6. The following tables are actual unit sales for six months and an initial forecast
for January.
a. Calculate the forecasts for the remaining five months using exponential smoothing.
simple conα= 0.2.
b. Calculate the MAD of the forecasts.
Real Forecasted
January 100 80
 February 94
March 106
April 80
May 68
June 94

Zeus Computer Chips, Inc. had important contracts to produce microprocessors.

Centrino type. The market declined in the three previous years due to dual-core modules,
that Zeus does not produce, so he has the tedious task of forecasting the coming year. The work
It is unpleasant because the company has not been able to find substitutes for its lines of
products. Here is the demand for the past 12 quarters:
2007 2008 2009
I 4800 I 3500 I 3200
II 3500 II 2700 II 2100
III 4300 III 3500 III 2700
IV 3000 IV 2400 IV 1700
Using the decomposition technique, forecast the four quarters of 2010.
8. The sales data for two years is as follows. The data is aggregated with two
sales months in each "period".
Months Sales Months Sales
January-February 109 January-February 115
March-April 104 March-April 112
May-June 150 May-June 159
July-August 170 July-August 182
September-October 120 September-October 126
November November
100 106
December December
a) Plot the data.
b) Compose a simple linear regression model for the sales data.
c) In addition to the regression model, determine the multiplier factors of the index.
seasonal. It is assumed that a complete cycle is one year.
d) With the results from sections b) and c), prepare a forecast for the upcoming year.
9. The tracking signals calculated with the history of the past demand of three
The products are as follows. Each product uses the same forecasting technique.
SS1 SS2 SS3
1 -2.70 1.54 0.10
2 -2.32 −0.64 0.43
3 -1.70 2.05 1.08
4 -1.10 2.58 1.74
5 -0.87 -0.95 1.94
6 -0.05 -1.23 2.24
7 0.10 0.75 2.96
8 0.40 -1.59 3.02
9 1.50 0.47 3.54
 10 2.20 2.74 3.75

Comment on the follow-up signals of each product and indicate their implications.
10. The following table shows the previous two years of sales information.
quarterly. Assume that there are trends and seasonal factors, and that the seasonal cycle
It is one year. With time series decomposition, I forecast the quarterly sales.
of the following year.
Quarter Sales Quarter Sales
1 160 5 215
2 195 6 240
3 150 7 205
4 140 8 190
Tucson Machinery, Inc. manufactures numerically controlled machines that are sold at a price
average of 0.5 million dollars each. Sales of these machines over the two years
previous ones are:
Trimestre Cantidad(unidades) Trimestre Cantidad(unidades)
2008 2009
I 12 I 16
II 18 II 24
III 26 III 28
IV 16 IV 18
a) Draw a straight line by hand (or make a regression with Excel).
b) Find the trend and seasonal factors.
c) Forecast the sales for 2010.
Not all the items in your stationery store are distributed.
uniformly concerning the demand, so he decides to forecast the demand
to plan your assortment. The past data from usual account notebooks for August,
they are the following:
Week 1300 Week 3600
Week 2400 Week 4700
a. With a three-week moving average, what would your forecast be for the upcoming week?
b. With exponential smoothing and = 0.20, if the exponential forecast for week 3 was calculated
as the average of the first two weeks [(300 + 400)/2 = 350], what would be its
forecast for week 5?
12. The actual tabulated demand for an item during a period of nine is given below.
months (January to September). Their supervisor wants to test two forecasting methods for
see which one turned out better during the period.
Month Real Month Real
January 110 June 180
February 130 July 140
March 150 August 130
April 170 September 140
May 160
a) Forecast from April to September with a three-month moving average.
b) Using simple exponential smoothing with an alpha of 0.3, calculate from April to September.
 c) With the DAM decide which method produced the best forecast over the six-month period.
13. A certain forecasting model was applied to anticipate a six-month period. Here are
the forecasted and actual demand:
Real Forecast
April 250 200
May 325 250
June 400 325
Julio 350 300
August 375 325
September 450 400
Find the tracking signal and indicate whether you think the model used provides acceptable answers.
14. Harlen Industries has a simple forecasting model: it takes the actual demand of the same
month of the previous year and is divided by the fractional number of weeks in that month. This gives
the average weekly demand for the month. With this weekly average, the forecasts are made
weeks of the same month of this year. The technique was used to forecast eight weeks of
this year, which are shown below along with the actual demand. The following eight
weeks show the forecast (based on last year) and the actual demand:
Demand Demand Demand Demand
Week Week
predicted real forecasted real
1 140 137 5 140 180
2 140 133 6 150 170
3 140 150 7 150 185
4 140 160 8 150 205
a) Calculate the DAM of the forecast errors.
b) With the SCEP, calculate the tracking signal.
c) Based on your responses in a) and b), comment on the forecasting method
Harlen.
15. La tabla siguiente contiene la demanda de los últimos 10 meses.
Month Demandareal My Demand area
1 31 6 36
2 34 7 38
3 33 8 40
4 35 9 40
5 37 10 41
a) Calculate the forecast using simple exponential smoothing for these data with an α of .30 and
an initial forecast (F1) of 31.
b) Calculate the exponential smoothing forecast with trend for these data, with a
of .30, a preliminary trend forecast (T1) of 1 and a smoothed forecast
initial exponential (F1) of 30.
c) Calculate the mean absolute deviation (MAD) of each forecast. Which one is the best?

16. In this problem, you will test the validity of your forecasting model. Next
The forecasts from the model used and the actual demand are given.
 Week Forecast Real
1 800 900
2 850 1000
3 950 1050
4 950 900
5 1000 900
6 975 1100
Using the method established in the text, calculate the DAM and the tracking signal. Then decide if
the forecasting model used provides reasonable results.
17. Suppose that your merchandise inventory for sale is maintained on the basis of the
forecasted demand. If the distributor's sales staff calls on the first day of
each month, calculate your sales forecast using the three methods requested here.
Real
June 140
July 180
August 170
a) With a three-month simple moving average, what is the forecast for September?
b) With a weighted moving average, what is the forecast for September with values
relatives of .20, .30 and .50 for June, July and August, respectively?
c) Through simple exponential smoothing, and assuming that the forecast for June was
130, forecast the sales for September with a smoothing constant α of .30.
18. The historical demand for a product is as follows:
Demand
April 60
May 55
June 75
July 60
August 80
September 75
a. With a simple moving average of four months, calculate a forecast for October.
b. Using simple exponential smoothing with α = 0.2 and a forecast for September
= 65, calculate a forecast for October.
c. Using simple linear regression, calculate the trend line of the data
historical. On the X axis, April = 1, May = 2, and so on, while
that on the Y axis is the demand.
d. Calculate a forecast for October.
19. The following table shows the predicted demand for a product with its method
particular of forecast, along with the actual demand.
Forecast Real
1500 1550
1400 1500
1700 1600
1750 1650
1800 1700
a) Calculate the tracking signal with the mean absolute deviation and the continuous sum of
forecasting errors.
b) Comment if your forecasting method gives good predictions.
20. Your manager is trying to determine which forecasting method to use. Based on the following
historical data, calculate the following forecast and specify which procedure you would use.
Demand Demand
Month Month
real real
1 62 7 76
2 65 8 78
3 67 9 78
4 68 10 80
5 71 11 84
6 73 12 85
a) Calculate the simple three-month moving average forecast for periods 4 to 12.
b) Calculate the weighted moving average for three months with weights of 0.50, 0.30, and 0.20.
for periods 4 to 12.
c) Calculate the simple exponential smoothing forecast for periods 2 to 12 with a
initial forecast (F1) of 61 and one of 0.30.
d) Calculate the exponential smoothing forecast with a trend component for the
Periods 2 to 12 with an initial trend forecast (T1) of 1.8, a forecast of
exponential smoothing initial (F1) of 60, one of 0.30 and one δ of 0.30.
e) Calculate the mean absolute deviation (MAD) of the forecasts made with each technique in
Periods 4 to 12. Which forecasting method do you prefer?
21. Perform a regression analysis on the demand without seasonal factors to forecast.
the demand in the summer of 2010, given the following historical demand data.
Year Demandareal Station
2008 Spring 205
Summer 140
Autumn 375
Winter 575
2009 Spring 475
Summer 275
Autumn 685
Winter 965
22. The actual demand for a product in the previous three months was
Three months ago 400 units
Two months ago 350 units
Last month 325 units
a) Using a simple three-month moving average, make a forecast for this month.
b) If the actual demand this month was 300 units, what would be your forecast for the month?
starter?
c) With simple exponential smoothing, what would your forecast be for this month if the forecast
exponential smoothing from three months ago would have been 450 units and the constant of
smoothing of 0.20?
23. After applying your forecasting model for six months, you decide to test it with DAM.
and a tracking signal. What follows is the forecast and actual demand for the period of
six months
Period Forecast Real
May 450 500
June 500 550
July 550 400
August 600 500
September 650 675
October 700 600
a) Find the tracking signal.
b) Decide if your forecasting routine is acceptable.
24. The earnings per share of two companies are recorded below, by quarter, of
first quarter of 2006 to the second of 2009. Forecast earnings per share for the
the rest of 2009 and 2010. With exponential smoothing, forecast the third period of 2009.
and with the time series decomposition method, the last two quarters of 2009
and the four quarters of 2010. It is much easier to solve the problem on a sheet of
computerized calculation to see what happens.
Earnings per share
Quarter Company A Company B
2006 I $1.67 $0.17
II 2.35 0.24
III 1.11 0.26
IV 1.15 0.34
2007 I 1.56 0.25
II 2.04 0.37
III 1.14 0.36
IV 0.38 0.44
2008 I 0.29 0.33
II -0.18 (loss) 0.40
III -0.97 (loss) 0.41
IV 0.20 0.47
2009 I -1.54 (loss) 0.30
II 0.38 0.47
a) For the exponential smoothing method, take the first quarter of 2006 as
pronóstico inicial. Haga dos pronósticos: uno con = 0.10 y otro con = 0.30.
b) With the DAM method to check the performance of the forecasting model, plus the
Real data from 2006 to the second quarter of 2009, how well did the model perform?
c) With the decomposition of the time series forecasting method, forecast the
earnings per share for the last two quarters of 2009 and the four quarters of
2010. Is there any seasonal factor in the earnings?
d) With your forecasts, comment on each company.
25. Below are the revenues from sales of a public utility company
large from 1999 to 2009. Forecast the revenues from 2010 to 2013. Use your good judgment,
intuition or common sense to choose your model or method, and your data period.
Revenue (millions) Ingresos(millones)
1999 $4865.9 2005 $5094.4
2000 5067.4 2006 5108.8
2001 5515.6 2007 5550.6
2002 5728.8 2008 5738.9
2003 5497.7 2009 5860.0
2004 5197.7
26. Mark Price, the new production manager of Speakers and Company, has to find out
What variable most affects the demand for your line of stereo speakers? Not sure
that the unit price of the product or the effects of greater marketing are the main ones
sales drivers, and wants to apply a regression analysis to find out which factor
drives more demand in its market. The relevant information was collected in a
extensive marketing project that extended over the past 10 years and was poured into the
following data:
Sales/unit Advertising (thousands
Year Precio/unidad
(thousands) dollars)
1998 400 280 600
1999 700 215 835
2000 900 211 1100
2001 1300 210 1400
2002 1150 215 1200
2003 1200 200 1300
2004 900 225 900
2005 1100 207 1100
2006 980 220 700
2007 1234 211 900
2008 925 227 700
2009 800 245 690
a) Perform a regression analysis in Excel based on these data. With your results, answer.
the following questions.
b) Which variable, price or advertising, has a greater effect on sales, and how do you know?
c) Forecast the average annual sales of speakers from Speakers and Company based on
the regression results if the price was 300 dollars per unit and the amount spent
in advertising (in thousands) was 900 dollars.
27. Suppose an initial Ft of 300 units, a trend of 8 units, an alpha of 0.30 and a
delta of 0.40. If the actual demand was 288, calculate the forecast for the next period.
28. The following table contains the number of complaints received in a department store
during the first six months of operation.
Me Complaints Me Complaints
January 36 April 90
February 45 May 108
 March 81 June 144
If a three-month moving average were used, what would have been the forecast for May?
29. The following is the number of cases of merlot wine sold at the Connor Owen winery in
a period of eight years.
Boxes of Boxes of
Year Year
Merlot wine merlot wine
2002 270 2006 358
2003 356 2007 500
2004 398 2008 410
2005 456 2009 376
Estimate the calculated uniformity value at the end of 2009 with an exponential smoothing model.
and an alpha value of 0.20. Use the average demand from 2002 to 2004 as the initial forecast and
continue to smooth the forecast until 2009.