Forecasting and Predictive Analytics With Forecast X 7Th Edition Keating Solutions Manual Full Chapter PDF

Forecasting and Predictive Analytics
with Forecast X 7th Edition Keating

Solutions Manual
Visit to download the full and correct content document: https://testbankdeal.com/dow
nload/forecasting-and-predictive-analytics-with-forecast-x-7th-edition-keating-solution
s-manual/
Solutions to Chapter 6 Exercises
1. Explain the similarity between how time-series decomposition and Winters’ exponential
smoothing deal with seasonality.
The similarity with which Winters’ (Holt Winters’) and time series decomposition
deal with seasonality is that both methods deal with seasonality in a multiplicative
manner. Each method gives you an index of how each period (month or quarterly
are most common) compares to the average period for the year. An index above
one indicates a high period while an index below one indicated a low period.
These can be interpreted as percentages. For example a December index of 1.28
tells you that on average December is 28% above the norm for the year. An index
of 0.78 says that that period is only 78% of the average for the year. In either
model (Winters’ or time series decomposition) the sum of the seasonal indices
will equal the number of periods represented per year in the data (four if using
quarterly data and 12 if using monthly data).
2. Discuss the trend, the seasonal, and the cyclical components.
The trend is the long term movement in the data. If the overall movement is from
low to high there is a positive trend and if the overall movement is from high to
low there is a negative trend.
The seasonal part of time series decomposition measures the regular pattern in the
data for specific periods year after year. For example for most retailers the fourth
quarter (or maybe December) typically has greater than average sales and thus the
corresponding seasonal indices are above one, and usually of similar values (but
not always the same value).
The cyclical component measures very long term movements of the data that are
often thought of as waves. A complete cycle from trough to trough (or peak to
peak) might take many years. In business the length and amplitude of cycles is
typically inconsistent.
3. What is the difference between seasonal factors and seasonal indices?
Seasonal factors are calculated for each period. Thus, there is a seasonal factor for
every December (or every fourth quarter). If you have twenty years of historical
data you would have twenty seasonal factors for each month or quarter. A
seasonal index is an average of all the seasonal factors for any specific period.
4. How is the long-term trend determined for a time-series decomposition model?
The long term trend in the classical time series decomposition model, as discussed
in the text, is estimated by using an OLS time trend with the centered moving
average as the dependent variable and a time index as the independent variable.
5. How do true cycles and the cycles typically found in business data differ?
Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution
without the prior written consent of McGraw-Hill Education.
A true cycle has a constant amplitude and a constant periodicity. In a time series
graph the amplitude is the vertical distance from trough to peak (or peak to
trough). The periodicity is the length of time (horizontal distance) from peak to
peak (or trough to trough). With business data such consistency does not exist.
For this reason some prefer to use the term business fluctuations since the
periodicity and amplitude vary along the time series of data.
6. Using your own words, write a description of each of the four components of the classic
time series decomposition technique. Avoid using mathematical relationships and
technical jargon as much as possible so that your explanations can be understood by
almost anyone.
The four components of time series decomposition are: 1) the trend; 2)
seasonality; 3) the cycle; and 4) random (irregular) fluctuations.
The trend is estimated form data that has been deseasonalized (the seasonality has
been removed). This is accomplished by using moving averages that include
every period of the year.
Seasonality is measured by comparing the actual value for any period with the
deseasonalized value for that period. The way in which the actual and
deseasonalized values are compared is by dividing the actual by the
deseasonalized value for each period. If the actual is greater than the
deseasonalized value it is a high period so the seasonal factor is greater than one.
If the actual is less than the deseasonalized value it is a low period and the
seasonal factor is less than one.
The cyclical value for each period is found by comparing the deseasonalized
value for each period with the trend value for each period. This is done by
dividing the deseasonalized value by the trend value. When this ratio is greater
than one the cycle is high and when below one the cycle is low.
The random (or irregular) part is the noise in the data caused by factors that can
not be forecast because of their random nature. In practice the random part of the
model is usually assumed to be one.
7. Suppose that sales of a household appliance are reported to be 13,000 units during the first
quarter of the year. The seasonal index for the first quarter is 1.24. Use this information to
make a forecast of sales for the entire year. Actual sales for the year were 42,000 units.
Calculate your percentage error for the year. What percentage error would result if you
forecast sales for the year by simply multiplying the 13,000 units for the first quarter by 4?
With actual sales of 13,000 and a seasonal index of 1.24 the deseasonalized value
would be: 13,000/1.24 which is 10,483.87. This represents the “typical” average
sales per quarter. Multiplying this by 4 (the number of quarters) gives a forecast
for the year of: 10,483.87 x 4 = 41,935.48.

The percentage error would be: ((42,000 - 41,935.48)/42,000)*100 = 0.154%.
By simply multiplying actual sales (13,000) by 4 we would forecast sales to be
52,000. The resulting percentage error would be: ((42,000 – 52,000)/42,000)*100
= - 23.81%.
8. In a time-series decomposition of sales (in millions of units), the following trend has been
estimated:
CMAT = 2.315 + 0.196(T)
The seasonal indices have been found to be:
Quarter Seasonal Index
1 1.24
2 1.01
3 0.76
4 0.99
For the coming year, the time index and cycle factors are:
Quarter T CF
1 21 1.01
2 22 1.04
3 23 1.06
4 24 1.04
a. From this information, prepare a forecast for each quarter of the coming year.
Forecast
Forecast Seasonal =
Period Trend Index Cycle B*C*D
21 16.431 1.27 1.01 21.076
22 16.627 1.02 1.04 17.638
23 16.823 0.73 1.06 13.018
24 17.019 0.98 1.04 17.346

b. Actual sales for the year you forecast in part (a) were 17.2, 13.2, 10.8, and 14.2 for
quarters 1, 2, 3, and 4, respectively. Use these actual sales figures along with your
forecasts to calculate the mean absolute percentage error for the forecast period.
Absolute Absolute
Period Actual Forecast Error Error % Error
21 17.2 21.076 -3.876 3.876 22.535
22 13.2 17.638 -4.438 4.438 33.621
23 10.8 13.018 -2.218 2.218 20.534
24 14.2 17.346 -3.146 3.146 22.153
MAPE = 24.7
9. A tanning parlor located in a major shopping center near a large New England city has the
following history of customers over the last four years (data are in hundreds of customers
and months are the midmonth of each quarter): (c6p9)
Yearly
Year Feb May Aug Nov
Totals
2012 3.5 2.9 2.0 3.2 11.6
2013 4.1 3.4 2.9 3.6 14.0
2014 5.2 4.5 3.1 4.5 17.3
2015 6.1 5.0 4.4 6.0 21.5
a. Construct a table in which you show the actual data (given in the table), the centered
moving average, the centered moving-average trend, the seasonal factors, and the cycle
factors for every quarter for which they can be calculated in years 2012 through 2015.
Date Customers MA4 CMA CMAT SF Cf Time

Feb-12 3.5 1
May-12 2.9 2
Aug-12 2.0 2.90 2.98 2.90 0.67 1.03 3
Nov-12 3.2 3.05 3.11 3.10 1.03 1.00 4
Feb-13 4.1 3.18 3.29 3.29 1.25 1.00 5
May-13 3.4 3.40 3.45 3.49 0.99 0.99 6
Aug-13 2.9 3.50 3.64 3.68 0.80 0.99 7
Nov-13 3.6 3.78 3.91 3.88 0.92 1.01 8
Feb-14 5.2 4.05 4.08 4.08 1.28 1.00 9
May-14 4.5 4.10 4.21 4.27 1.07 0.99 10
Aug-14 3.1 4.33 4.44 4.47 0.70 0.99 11
Nov-14 4.5 4.55 4.61 4.66 0.98 0.99 12
Feb-15 6.1 4.68 4.84 4.86 1.26 1.00 13
May-15 5.0 5.00 5.19 5.05 0.96 1.03 14
Aug-15 4.4 5.38 5.25 15
Nov-15 6.0 5.45 16
b. Look at the seasonal index for each quarter as calculated in ForecastX™. Do they make
sense to you? Explain why or why not.
ForecastX
Seasonal
Date SF Indices
Feb-12 1.27
May-12 1.01
Aug-12 0.67 0.73
Nov-12 1.03 0.98
Feb-13 1.25 1.27
May-13 0.99 1.01
Aug-13 0.80 0.73
Nov-13 0.92 0.98
Feb-14 1.28 1.27
May-14 1.07 1.01
Aug-14 0.70 0.73
Nov-14 0.98 0.98
Feb-15 1.26 1.27
May-15 0.96 1.01
Aug-15 0.73
Nov-15 0.98
For the three seasonal factors calculated for the first quarter (February) we have
1.25, 1.28 and 1.26. ForecastX calculates the seasonal index for February to be
1.27. This illustrates the consistency between manual calculations (in Excel) and
ForecastX. ForecastX calculates the average for each quarter and then normalizes
the results to add to four. If you add any four consecutive seasonal indices from
ForecastX you will find that they add to four. The other three seasonal indices are:
1.01, 0.73, and 0.98.
c. Make a forecast of the number of customers for each quarter of 2016.
The forecast is the product of CMAT*SI*CF as shown below. The cycle factors
(CF) were taken from ForecastX estimates.

Date Customers CMAT SI Cf Forecast
Feb-16 ? 5.64 1.27 1.01 7.25
May-16 ? 5.84 1.01 1.01 5.98
Aug-16 ? 6.03 0.73 1.01 4.44
Nov-16 ? 6.23 0.98 1.01 6.19
d. The actual numbers of customers served per quarter in 2016 were 6.8, 5.1, 4.7, and 6.5
for
quarters 1 Absolute Absolute
through 4, Date Customers Forecast Error Error % Error
Feb-16 6.8 7.25 -0.45 0.45 6.64
May-16 5.1 5.98 -0.88 0.88 17.31
Aug-16 4.7 4.44 0.26 0.26 5.47
Nov-16 6.5 6.19 0.31 0.31 4.83
MAPE = 8.56
respectively (numbers are in hundreds). Calculate the MAPE for 2016.
e. Using the results provided in the tables produced by ForecastX™, prepare a time-series
plot of the actual data, the centered moving averages, the long-term trend, and the values
predicted by your model for 2012 through 2016 (where data are available).

Acual, CMA, CMAT & Forecast
8.00
6.00
4.00
2.00
0.00
Nov-2015
May-2012
Nov-2012
May-2013
Nov-2013
May-2014
Nov-2014
May-2015
May-2016
Nov-2016
Aug-2012
Aug-2013
Feb-2014
Aug-2014
Aug-2015
Aug-2016
Feb-2012
Feb-2013
Feb-2015
Feb-2016
Actual Centered Moving Average
CMA Trend Forecasted Data
10. Carl Lipke is the marketing VP for a propane gas distributor. He would like to have a
forecast of sales on a quarterly basis, and he has asked you to prepare a time-series
decomposition model. The data for 2005 through 2016 follow: (c6p10)
Propane Gas Sales in Millions of Pounds (Total at End-Month of Each Quarter)
Year March June September December
2005 6.44 4.85 4.67 5.77
2006 6.22 4.25 4.14 5.34
2007 6.07 4.36 4.07 5.84
2008 6.06 4.24 4.20 5.43
2009 6.56 4.25 3.92 5.26
2010 6.65 4.42 4.09 5.51
2011 6.61 4.25 3.98 5.55
2012 6.24 4.34 4.00 5.36
2013 6.40 3.84 3.53 4.74
2014 5.37 3.57 3.32 5.09
2015 6.03 3.98 3.57 4.92
2016 6.16 3.79 3.39 4.51

a. To help Carl Lipke see how propane gas sales have varied over the 12-year period,
prepare a time-series plot of the raw data and the deseasonalized data (i.e., the centered
moving averages).
Propane Gas Sales (Millions of

Pounds)
7
6
5
4
3
2
1
0
Mar-12
Mar-05
Sep-05
Mar-06
Sep-06
Mar-07
Sep-07
Mar-08
Sep-08
Mar-09
Sep-09
Mar-10
Sep-10
Mar-11
Sep-11
Sep-12
Mar-13
Sep-13
Mar-14
Sep-14
Mar-15
Sep-15
Mar-16
Sep-16
Sales Centered Moving Average
b. Use ForecastX™ to find seasonal indices for quarters 1 through 4. Write a short
paragraph in which you explain to Carl Lipke exactly what these indices mean.
Seasonal
Indices
1.27
0.85
0.80
1.09
The first quarter seasonal index of 1.27 tells Mr. Lipke that the first
quarter is a high quarter during which sales are typically 1.27 times the
quarterly average for the year. This quarter includes January February and
March which are the coldest months of the year so this result is quite
logical.
The second quarter of April, May and June may also see some cool
weather but the index of 0.85 tells us that this quarter is well below
average (about 85% of average). The same is true of the third quarter
which includes July, August and September. During the second and third
quarters the main demand would be for grilling and camping which would
require much less propane than for home heating.
In the months of October, November and December home heating demand
starts to build as indicated by the seasonal index of 1.09.
c. Plot the values of actual sales, the centered moving averages, and the trend. All of these
can be found in your ForecastX™ results.
Actual, Centered Moving Average and Trend

Values
7.00
6.00
5.00
4.00
3.00
2.00
1.00
0.00
Sep-2008
Sep-2016
Mar-2005
Sep-2005
Mar-2006
Sep-2006
Mar-2007
Sep-2007
Mar-2008
Mar-2009
Sep-2009
Mar-2010
Sep-2010
Mar-2011
Sep-2011
Mar-2012
Sep-2012
Mar-2013
Sep-2013
Mar-2014
Sep-2014
Mar-2015
Sep-2015
Mar-2016
Actual Sales Centered Moving Average CMA Trend
d. From ForecastX™, get a forecast for 2017Q1 through 2017Q4 based on the time-series
decomposition model. Enter your forecast values into an Excel sheet that you set up like
in the table shown below. Given the actual values shown in the table, calculate the mean
absolute percentage error (MAPE) for 2017.
Actual Forecast Absolute Absolute

Date Sales Sales Error Error % Error
Mar-17 5.39 5.72 -0.33 0.33 6.10
Jun-17 3.56 3.79 -0.23 0.23 6.43
Sep-17 3.03 3.58 -0.55 0.55 18.00
Dec-17 4.03 4.84 -0.81 0.81 20.04
MAPE = 12.64
Date Actual Sales Forecast Sales Error Absolute Error Absolute % Error
Mar-17 5.39
Jun-17 3.56
Sep-17 3.03
Dec-17 4.03

MAPE =
11. The Bechtal Tire Company (BTC) is a supplier of automotive tires for U.S. car companies.
BTC has hired you to analyze its sales. For this problem, do all the work in ForecastX™
and be sure to request the MAPE in the statistics tab. Data from 1995Q1 through 2016Q4
are given in the following table (in thousands of units): (c6p11)
Sales Sales Sales Sales Sales

of of of of of
Period Tires Period Tires Period Tires Period Tires Period Tires
Mar-95 2,029 Mar-00 1,752 Mar-05 2,155 Mar-10 1,994 Mar-14 1,914
Jun-95 2,347 Jun-00 2,138 Jun-05 2,231 Jun-10 2,251 Jun-14 2,076
Sep-95 1,926 Sep-00 1,927 Sep-05 1,971 Sep-10 1,855 Sep-14 1,787
Dec-95 2,162 Dec-00 1,546 Dec-05 1,875 Dec-10 1,852 Dec-14 1,763
Mar-98 2,064 Mar-03 2,128 Mar-08 1,401 Mar-13 1,669 Mar-17
Jun-98 2,517 Jun-03 2,538 Jun-08 1,535 Jun-13 1,973 Jun-17
Sep-98 2,147 Sep-03 2,081 Sep-08 1,327 Sep-13 1,878 Sep-17
Dec-98 2,524 Dec-03 2,223 Dec-08 1,494 Dec-13 1,561 Dec-17
Mar-99 2,451 Mar-04 2,027 Mar-09 1,456
Jun-99 2,718 Jun-04 2,727 Jun-09 1,876
Sep-99 2,229 Sep-04 2,140 Sep-09 1,646
Dec-99 2,190 Dec-04 2,270 Dec-09 1,813
a. Write a report to Bechtal Tire Company in which you explain what a time-series
decomposition analysis shows about its tire sales. Include in your discussion seasonal,
cyclical, and trend components.

Sales of Tires
3,000
2,500
2,000
1,500
1,000
500
0
Mar-02
Mar-95
Mar-96
Mar-97
Mar-98
Mar-99
Mar-00
Mar-01
Mar-03
Mar-04
Mar-05
Mar-06
Mar-07
Mar-08
Mar-09
Mar-10
Mar-11
Mar-12
Mar-13
Mar-14
Mar-15
Mar-16
Trend and Cycle (CMA)
3,000
2,500
2,000
1,500
1,000
500
0
Mar-2002
Mar-1995
Mar-1996
Mar-1997
Mar-1998
Mar-1999
Mar-2000
Mar-2001
Mar-2003
Mar-2004
Mar-2005
Mar-2006
Mar-2007
Mar-2008
Mar-2009
Mar-2010
Mar-2011
Mar-2012
Mar-2013
Mar-2014
Mar-2015
Mar-2016
Centered Moving Average Trend
The top graph above allows us to see the clear seasonality in the data as shown by the
reasonably regular upward and downward movements of the series that repeat year after
year. Through analysis it turns out that the second quarter of each year (ending in June) is
the peak season for this companies tire sales. The seasonal indices are:
Seasonal
Indices
0.97
1.10
0.98
0.96
The lower graph shows that over this period there is a long term negative trend to sales as
indicated by the dashed line. The dotted line shows that sales are cyclical in nature with
wave like movement below and above the long term trend.
b. Show a time-series graph with the actual data and the values that the time-series
decomposition model would predict for each quarter from 1995Q1 through 2017Q4
(some data will be missing for certain historical quarters, and, of course for 2017, you
will have only the forecast values).
Actual and Forecasted Values
3,000
2,500
2,000
1,500
1,000
500
0
Mar-1995
Mar-1997
Mar-1996
Mar-1998
Mar-1999
Mar-2000
Mar-2001
Mar-2002
Mar-2003
Mar-2004
Mar-2005
Mar-2006
Mar-2007
Mar-2008
Mar-2009
Mar-2010
Mar-2011
Mar-2012
Mar-2013
Mar-2014
Mar-2015
Mar-2016
Mar-2017
Actual Forecasted
c. If actual sales for 2017 were Q1 = 1,445.1, Q2 = 1,683.8, Q3 = 1,586.6, and Q4 =

1,421.3, what MAPE would result from your 2017 forecast?
Absolute Absolute
Date Actual Forecast Error Error % Error
Mar-2017 1,445.10 1,650.21 -205.11 205.11 14.19
Jun-2017 1,683.80 1,871.53 -187.73 187.73 11.15
Sep-2017 1,586.60 1,662.02 -75.42 75.42 4.75
Dec-2017 1,421.30 1,626.48 -205.18 205.18 14.44
MAPE = 11.13
How does this MAPE compare to the MAPE ForecatsX™ calculated for the historic
period?
MAPE = 5.38%

The MAPE for the historic period from ForecastX is 5.38%. It is common for the
MAPE to be lower in sample (the historic period) than out of sample (the forecast
period).
12. A regional supplier of jet fuel is interested in forecasting its sales. These sales data are
shown for the period from 2002Q1 to 2017Q4 (data in billions of gallons):
(c6p12)
Jet Fuel Sales (Billions of Gallons)
Year Q1 Q2 Q3 Q4
2002 23.86 23.97 29.23 24.32
2003 23.89 26.84 29.36 26.30
2004 27.09 29.42 32.43 29.17
2005 28.86 32.10 34.82 30.48
2006 30.87 33.75 35.11 30.00
2007 29.95 32.63 36.78 32.34
2008 33.63 36.97 39.71 34.96
2009 35.78 38.59 42.96 39.27
2010 40.77 45.31 51.45 45.13
2011 48.13 50.35 56.73 48.83
2012 49.02 50.73 53.74 46.38
2013 46.32 51.65 52.73 47.45
2014 49.01 53.99 55.63 50.04
2015 54.77 56.89 57.82 53.30
2016 54.69 60.88 63.59 59.46
2017 61.59 68.75 71.33 64.88
a. Prepare a time series graph of these data. What, if any, seasonal pattern do you see in the
plot? Explain.

Jet Fuel Sales (Billions of Gallons
80
70
60
50
40
30
20
10
0
Jul-03
Jul-05
Jul-07
Jul-09
Jul-11
Jul-13
Jul-15
Jul-17
Nov-02
Nov-04
Nov-06
Nov-08
Nov-10
Nov-12
Nov-14
Nov-16
Mar-02
Mar-04
Mar-06
Mar-08
Mar-10
Mar-12
Mar-14
Mar-16
The data show an upward trend and seasonality. There also appears to be a drop in
sales starting in late 2011 then sales pick up again starting in 2013.
b. Use ForecastX™ to make a time series decomposition (TSD) forecast for 2018. Write a
brief report explaining your forecast. Include a graph of the fitted values, the forecast
values, and the actual sales.
Jet Fuel Sales TSD Model

80
60
40
20
0
Nov-2002
Jul-2003
Nov-2004
Jul-2005
Nov-2006
Jul-2007
Nov-2008
Jul-2009
Nov-2010
Jul-2011
Nov-2012
Jul-2013
Nov-2014
Jul-2015
Nov-2016
Jul-2017
Nov-2018
Mar-2002
Mar-2004
Mar-2006
Mar-2008
Mar-2010
Mar-2012
Mar-2014
Mar-2016
Mar-2018
Actual Forecast Fitted Values
We see the forecast as a dotted line which appears reasonable given the historic data. The fitted
values and the actual values are very difficult to tell apart in the graph. This is often the case with
time series decomposition because the reality is the process breaks the components apart,
forecasts each component then puts them back together. In the process of making the fimal
forecast we have: Sales = Trendx SeasonalxCyclical = (CMAT)x(Sales/CMA)x(CMA/CMAT).
From this we have: (CMAT)x(Sales/CMA)x(CMA/CMAT) = Sales. The only deviations would
be the random (irregular) component, which often is quite small.
c. Develop two other forecasts of jet fuel sales using the following methods:

1. A Winters' exponential smoothing model;
Jet Fuel Sales: Winters' Model

90
80
70
60
50
40
30
20
10
0
Mar-2002
Mar-2003
Mar-2004
Mar-2005
Mar-2006
Mar-2007
Mar-2008
Mar-2009
Mar-2010
Mar-2011
Mar-2012
Mar-2013
Mar-2014
Mar-2015
Mar-2016
Mar-2017
Mar-2018
Sep-2002
Sep-2003
Sep-2004
Sep-2005
Sep-2006
Sep-2007
Sep-2008
Sep-2009
Sep-2010
Sep-2011
Sep-2012
Sep-2013
Sep-2014
Sep-2015
Sep-2016
Sep-2017
Sep-2018
2. A regression model using just time and quarterly dummy variables.
Jet Fuel Sales: Regression Model

80
70
60
50
40
30
20
10
0
Oct-2002
Feb-2005
Apr-2006
Oct-2009
Feb-2012
Apr-2013
Oct-2016
Jul-2004
Nov-2006
Nov-2013
Mar-2002
May-2003
Jun-2007
Jan-2008
Jul-2011
Mar-2009
May-2010
Jun-2014
Jan-2015
Jul-2018
Mar-2016
May-2017
Sep-2005
Sep-2012
Aug-2008
Aug-2015
Dec-2003
Dec-2010
Dec-2017
Compare the MAPEs for the three models you have developed, and comment on what you like
or dislike about each of the three models for this application.
TSD MAPE = 1.34%
Winters' MAPE = 2.67%
Regression MAPE = 5.05%

The time series decomposition model has the lowest in sample MAPE but that can be an
unreliable measure due to the way the model works. The out of sample MAPE for TSD
can often deteriorate quite a bit due to the difficulty of predicting the cycle factor. All
things considered many would say the Winters’ model is probably the best to use in this
case.
13. The following table contains quarterly data on Upper Midwest car sales (CS) in thousands
for 1996Q1 through 2016Q4:
Upper Midwest Car Sales (CS) (c6p13)
Year Q1 Q2 Q3 Q4
1996 407.6 431.5 441.6 306.2
1997 328.7 381.3 422.6 369.4
1998 456.3 624.3 557.5 436.7
1999 485.0 564.3 538.3 412.5
2000 555.0 682.7 581.3 509.7
2001 662.7 591.1 616.9 529.7
2002 641.2 632.7 576.6 475.0
2003 542.8 558.9 581.7 537.8
2004 588.1 626.5 590.9 580.1
2005 589.2 643.2 593.9 612.2
2006 586.1 699.4 734.4 753.8
2007 691.6 793.4 864.9 840.8
2008 653.9 754.8 883.6 797.7
2009 722.2 788.6 769.9 725.5
2010 629.3 738.6 732.0 598.8
2011 603.9 653.6 606.1 539.7
2012 461.3 548.0 548.4 480.4
2013 476.6 528.2 480.4 452.6
2014 407.2 498.5 474.3 403.7
2015 418.6 470.2 470.7 375.7
2016 371.1 425.5 397.3 313.5
a. Prepare a time-series plot of Upper Midwest car sales from 1996Q1 through 2016Q4.

Upper Midwest Car Sales (000)
1000
900
800
700
600
500
400
300
200
100
0
Nov-99
Jul-03
Jul-14
Jun-04
May-05
Apr-06
Jun-15
Aug-02
Nov-10
May-16
Dec-09
Aug-13
Mar-96
Feb-97
Jan-98
Dec-98
Sep-01
Mar-07
Feb-08
Jan-09
Sep-12
Oct-00
Oct-11
b. Use ForecastX™ to do a time-series decomposition forecast for 2017 (be sure to request
the MAPE). In the results, you see the seasonal indices. Do they make sense? Why or
why not?
Upper Midwest Car Sales (000)

1,000
800
600
400
200
0
Jun-1998
Jun-2001
Jun-2004
Jun-2007
Jun-2010
Jun-2013
Jun-2016
Mar-1996
Sep-1997
Mar-1999
Sep-2000
Mar-2002
Sep-2003
Mar-2005
Sep-2006
Mar-2008
Sep-2009
Mar-2011
Sep-2012
Mar-2014
Sep-2015
Mar-2017
Dec-1996
Dec-1999
Dec-2002
Dec-2005
Dec-2008
Dec-2011
Dec-2014
Dec-2017
The MAPE for this model during the historic period is 5.03%.
Quarter
Ending Seasonal
Month Indices
March 0.95
June 1.07
September 1.05
December 0.93
Spring and summer months are typically strong months for car sales. Part of this may be due to
new car releases as well as the desire to have a new vehicle for summer vacations.
c. ForecastX™ calculated the historic MAPE as a measure of fit. Write a short explanation
of what this MAPE means to a manager.
This MAPE tells a manager how well the model fit the historic data in a manner
that can be compared with national norms or with other industries. Positive errors
are not offset by negative ones because the absolute value of errors are used in the
calculation.
d. Now calculate the MAPE for the 2017Q1–2017Q4 forecast horizon as a measure of
accuracy, given that the actual values of CS for 2017 were:
2017Q1 301.1
2017Q2 336.7
2017Q3 341.8
2017Q4 293.5
Absolute Absolute
Date Actual Predicted Error Error % Error
Mar-17 301.1 380.13 -79.03 79.03 26.25
Jun-17 336.7 427.98 -91.28 91.28 27.11
Sep-17 341.8 424.65 -82.85 82.85 24.24
Dec-17 293.5 372.51 -79.01 79.01 26.92
MAPE = 26.13
e. Prepare a Winters’ exponential smoothing forecast of CS using data from 1996Q1

through 2016Q4 as the basis for a forecast of 2017Q1–2017Q4. Compare these results in
terms of fit and accuracy with the results from the time-series decomposition forecast.
The MAPE for this model during the historic period is 6.87% compared
with 5.03% for the time series decomposition model. This is “fit.”
Absolute Absolute
Date Actual Predicted Error Error % Error
Mar-17 301.1 333.44 -32.34 32.34 10.74
Jun-17 336.7 387.74 -51.04 51.04 15.16
Sep-17 341.8 374.49 -32.69 32.69 9.57
Dec-17 293.5 313.50 -20.00 20.00 6.81
MAPE = 10.57
However, we see that for the forecast period the MAPE for the Holt
Winters model is better, given how ForecastX predicted the cycle factor
for 2017. These MAPEs measure “accuracy.”
.

Another random document with
no related content on Scribd:
"Thirty-two."
"Twenty-five."
"Good heavens," I said, "I've only got seven."
There was a shout of laughter.
"Then you'd better begin," said somebody. "Read them out."
I coughed doubtfully, and began;
"Napoleon."
There was another shout of laughter.
"I am afraid we can't allow that."
"Why ever not?" I asked in amazement.
"Well, you'd hardly find him in a kitchen, would you?"
I took out a handkerchief and wiped my brow. "I don't want to find him
in a kitchen," I said nervously. "Why should I? As a matter of fact, he's
dead. I don't see what the kitchen's got to do with it. Kitchens begin with a
K."
"But the game is called 'Furnishing a Kitchen.' You have to make a list
of things beginning with N which you would find in a kitchen. You
understood that, didn't you?"
"Y-y-yes," I said. "Oh y-y-y-yes. Of course."
"So Napoleon——"
I pulled myself together with a great effort.
"You don't understand," I said with dignity. "The cook's name was
Napoleon."
"Cooks aren't called Napoleon," said everybody.
"This one was. Carrie Napoleon. Her mistress was just as surprised at
first as you were, but Carrie assured her that——"
"No, I'm afraid we can't allow it."
"I'm sorry," I said; "I'm wrong about that. Her name was Carrie Smith.
But her young man was a soldier, and she had bought a 'Life of Napoleon'
for a birthday present for him. It stood on the dresser—it did, really—
waiting for her next Sunday out."
"Oh! Oh, well, I suppose that is possible. Go on."
"Gnats," I went on nervously and hastily. "Of course I know that——"
"Gnats are spelt with a G," they shrieked.
"These weren't. They had lost the G when they were quite young, and
consequently couldn't bite at all, and cook said that——"
"No, I'm afraid not."
"I'm sorry," I said resignedly. "I had about forty of them—on the
dresser. If you won't allow any of them, it pulls me down a lot. Er—then we
have Nitro-glycerine."
There was another howl of derision.
"Not at all," I said haughtily. "Cook had chapped hands very badly, and
she went to the chemist's one evening for a little glycerine. The chemist was
out, and his assistant—a very nervous young fellow—gave her nitro-
glycerine by mistake. It stood on the dresser, it did, really."
"Well," said everybody very reluctantly, "I suppose——"
I went on hastily:
"That's two. Then Nobody. Of course, you might easily find nobody in
the kitchen. In fact you would pretty often, I should say. Three. The next is
Noon. It could be noon in the kitchen as well as anywhere else. Don't be
narrow-minded about that."
"All right. Go on."
"Non-sequitur," I said doubtfully.
"What on earth——"
"It's a little difficult to explain, but the idea is this. At most restaurants
you can get a second help of anything for half-price, and that is technically
called a 'follow.' Now, if they didn't give you a follow, that would be a Non-
sequitur.... You do see that, don't you?"
There was a deadly silence.
"Five," I said cheerfully. "The last is Notting Hill. I must confess," I said
magnanimously, "that I am a bit doubtful whether you would actually find
Notting Hill in a kitchen."
"You don't say so!"
"Yes. My feeling is that you would be more likely to find the kitchen in
Notting Hill. On the other hand, it is just possible that, as Calais was found
engraven on Mary's heart, so, supposing the cook died—— Oh, very well.
Then it remains at five."
Of course you think that, as I only had five, I came out last. But you are
wrong. There is a pleasing rule in this game that, if you have any word in
your list which somebody else has, you cannot count it. And as all the
others had the obvious things—such as a nutmeg-grater or a neck of mutton
or a nomelette—my five won easily. And you will note that, if only I had
been allowed to count my gnats, it would have been forty-five.
II. GETTING THE NEEDLE
He was a pale, enthusiastic young man of the name of Simms; and he

held forth to us at great length about his latest hobby.
"Now I'll just show you a little experiment," he wound up; "one that I
have never known to fail. First of all, I want you to hide a needle
somewhere, while I am out of the room. You must stick it where it can be
seen—on a chair—or on the floor if you like. Then I shall come back and
find it."
"Oh, Mr Simms!" we all said.
"Now, which one of you has the strongest will?"
We pushed Jack forward. Jack is at any rate a big man.
"Very well. I shall want you to take my hand when I come in, and look
steadily at the needle, concentrate all your thoughts on it. I, on the other
hand, shall make my mind a perfect blank. Then your thoughts will
gradually pass into my brain, and I shall feel myself as it were dragged in
the direction of the needle."
"And I shall feel myself, as it were, dragged after you?" said Jack.
"Yes; you mustn't put any strain on my arm at all. Let me go just where
I like, only will me to go in the right direction. Now then."
He took out his handkerchief, put it hastily back, and said: "First, I shall
want to borrow a handkerchief or something."
Well, we blindfolded him, and led him out of the room. Then Muriel got
a needle, which, after some discussion, was stuck into the back of the
Chesterfield. Simms returned, and took Jack's left hand.
They stood there together, Jack frowning earnestly at the needle, and
Simms swaying uncertainly at the knees. Suddenly his knees went in
altogether, and he made a little zig-zag dash across the room, as though he
were taking cover. Jack lumbered after him, instinctively bending his head
too. They were brought up by the piano, which Simms struck with great
force. We all laughed, and Jack apologised.
"You told me to let you go where you liked, you know," he said.
"Yes, yes," said Simms rather peevishly, "but you should have willed
me not to hit the piano."
As he spoke he tripped over a small stool, and, flinging out an arm to

save himself, swept two photograph frames off an occasional table.
"By Jove!" said Jack, "that's jolly good. I saw you were going to do that,
and I willed that the flower vase should be spared. I'm getting on."
"I think you had better start from the door again," I suggested. "Then
you can get a clear run."
They took up their original positions.
"You must think hard, please," said Simms again. "My mind is a perfect
blank, and yet I can feel nothing coming."
Jack made terrible faces at the needle. Then, without warning, Simms
flopped on to the floor at full length, pulling Jack after him.
"You mustn't mind if I do that," he said, getting up slowly.
"No," said Jack, dusting himself.
"I felt irresistibly compelled to go down," said Simms.
"So did I," said Jack.
"The needle is very often hidden in the floor, you see. You are sure you
are looking at it?"
They were in a corner with their backs to it; and Jack, after trying in
vain to get it over his right shoulder or his left, bent down and focussed it
between his legs. This must have connected the current; for Simms turned
right round and marched up to the needle.
"There!" he said triumphantly, taking off the bandage.
We all clapped, while Jack poured himself out a whisky. Simms turned
to him.
"You have a very strong will indeed," he said, "one of the strongest I
have met. Now, would one of the ladies like to try?"
"Oh, I'm sure I couldn't," said all the ladies.
"I should like to do it again," said Simms modestly. "Perhaps you, sir?"
"All right, I'll try," I said.
When Simms was outside I told them my idea.
"I'll hold the needle in my other hand," I said, "and then I can always
look at it easily, and it will always be in a different place, which ought to
muddle him."
We fetched him in, and he took my left hand....
"No, it's no good," he said at last, "I don't seem to get it. Let me try the
other hand...."
I had no time to warn him. He clasped the other hand firmly; and, from
the shriek that followed, it seemed that he got it. There ensued the "perfect
blank" that he had insisted on all the evening. Then he pulled off the
bandage, and showed a very angry face.
Well, we explained how accidental it was, and begged him to try again.
He refused rather sulkily.
Suddenly Jack said: "I believe I could do it blindfold. Miss Muriel, will
you look at the needle and see if you can will me?"
Simms bucked up a bit, and seemed keen on the idea. So Jack was
blindfolded, the needle hid, and Muriel took his hand.
"Now, is your mind a perfect blank?" said Simms to Jack.
"It always is," said Jack.
"Very well, then. You ought soon to feel in a dreamy state, as though
you were in another world. Miss Muriel, you must think only of the needle."
Jack held her hand tight, and looked most idiotically peaceful. After
three minutes Simms spoke again.
"Well?" he said eagerly.
"I've got the dreamy, other-world state perfectly," said Jack, and then he
gave at the knees just for the look of the thing.
"This is silly," said Muriel, trying to get her hand away.
Jack staggered violently, and gripped her hand again.
"Please, Miss Muriel," implored Simms. "I feel sure he is just going to
do it."
Jack staggered again, sawed the air with his disengaged hand, and then
turned right round and marched for the door, dragging Muriel behind him.
The door slammed after them....
* * * * * * * *
There is a little trick of sitting on a chair and picking a pin out of it with
the teeth. I started Simms—who was all eagerness to follow the pair, and
find out the mysterious force that was drawing them—upon this trick, for
Jack is one of my best friends. When Jack and Muriel came back from the
billiard-room, and announced that they were engaged, Simms was on his
back on the floor with the chair on the top of him—explaining, for the
fourth time, that if the thing had not overbalanced at the critical moment he
would have secured the object. There is much to be said for this view.
BACHELOR DAYS
THE BUTTER
"You mustn't think I am afraid of my housekeeper. Not at all. I

frequently meet her on the stairs, and give her some such order as "I think
—if you don't mind—I might have breakfast just a little later—er, yes,
about eight o'clock, yes, thank you." Or I ring the bell and say, "I—I—
want-my-boots." We both recognise that it is mine to command and hers to
obey. But in the matter of the butter I have let things slide, until the position
is rapidly becoming an untenable one. Yet I doubt if a man of imagination
and feeling could have acted otherwise, given the initial error. However,
you shall hear.
There are two sorts of butter, salt and fresh. Now, nobody is so fond of
butter as I am; but butter (as I have often told everybody) isn't butter at all
unless it is salt. The other kind is merely an inferior vaseline—the sort of
thing you put on the axles of locomotives. Imagine, then, my disgust, when
I took my first breakfast in these rooms eleven months ago, to find that the
housekeeper had provided me with a large lump of vaseline!
I hate waste in small things. Take care of the little extravagances, I say,
and the big ones will take care of themselves. My first thought on viewing
this pat of butter was, "It is difficult, but I will eat it." My second, "But I
must tell the housekeeper to get salt butter next time."
An ordinary-minded person would have stopped there. I went one

further. My third thought was this: "Housekeepers are forgetful creatures. If
I tell her now, she will never remember. Obviously I had better wait until
this pound is just finished and she is about to get in some more. Then will
be the time to speak." So I waited; and it was here that I made my mistake.
For it turned out that it was I who was the forgetful creature. And on the
fifteenth day I got up to find another large pound of vaseline on my table.
The next fortnight went by slowly. I kept my eye on every day, waiting for
the moment to come when I could say to the housekeeper, "You will be
getting me in some more butter this morning. Would you get salt, as I don't
much like the other?" Wednesday came, and there was just enough left for
two days. I would speak on the morrow.
But, alas! on the morrow there was another new pound waiting. I had
evidently misjudged the amount.
I forget what happened after that. I fancy I must have been very busy, so
that the question of butter escaped me altogether. Sometimes, too, I would
go away for a few days and the old butter would be thrown away and the
new butter bought, at a time when I had no opportunity of defending
myself. However it was, there came a time when I had been three months in
my rooms, and was still eating fresh butter; contentedly, to all appearances
—in the greatest anguish of soul, as it happened. And at the end of another
month I said: "Now, then, I really must do something about this."
But what could I do? After eating fresh butter for four months without
protest I couldn't possibly tell the housekeeper that I didn't like it, and
would she get salt in future. That would be too absurd. Fancy taking four
months to discover a little thing like that! Nor could I pretend that, though I
used to adore fresh butter, I had now grown tired of it. I hate instability of
character, and I could not lend myself to any such fickleness. I put it to you
that either of these courses would have shown deplorable weakness. No, an
explanation with the housekeeper was by that time impossible; and if
anything was to be done I must do it on my own responsibility. What about
buying a pound of salt butter myself, and feeding on it in secret? True, I
should have to get rid of a certain portion of fresh every day, but...
I don't know if you have ever tried to get rid of a certain portion of fresh
butter every day, when you are living in a flat at the very top of chambers in
London. Drop it out of the window once or twice, and it is an accident.
Three times, and it is a coincidence. Four times, and the policeman on duty
begins to think that there is more in it (if I may say so) than meets the eye....
But what about the fire, you will ask. Ah, yes; but I could foresee a day
when there would be no fire. One has to look ahead.
Besides, as I said, I hate waste. As any cook will tell you, the whole art
of housekeeping can be summed up in three words—Watch the butter.
More months passed and more pats of vaseline. Every day made an
explanation more hopeless. I had thoughts at one time of an anonymous
letter. Something in this style:
"MADAM,—One who is your friend says beware of vaseline. All is

discovered. Fly before it is too late. What is it makes the sea so salt? NaCl.
Sodium Chloride. THE BLACK HAND"
That would give her the impression, at any rate, that there were two
kinds of butter. Confound it all, by what right did she assume without
asking that I had a preference for fresh?
I have now been in my rooms nearly a year. Something must be done

soon. My breakfasts are becoming a farce. Meals which I used to enjoy I
now face as an ordeal. Is there to be no hope for me in the future?
Well, there is a chance? I shall have to wait until July; but with
something definite in view I am content to wait.
In July I hope to go to Switzerland for a month. Two days before

returning home I shall write to my housekeeper. Having announced the day
of my return, and given one or two instructions, I shall refer briefly to the
pleasant holiday which I have been enjoying. I shall remark perhaps on the
grandeur of the mountains and the smiling beauty of the valleys, I may
mention the area in square miles of the country....
And I shall dwell upon the habits of the native.

"... They live (I shall write) in extraordinary simplicity, chiefly upon the
products of their farms. Their butter is the most delightful I have ever tried.
It is a little salt to the taste, but after four weeks of it I begin to feel that I
shall never be able to do without salt butter again! No doubt, as made in
London it would be different from this, but I really think I must give it a
trial. So when you are ordering the things I mentioned for me, will you ask
for salt butter...."
And if that fails there remains only the one consolation. In three years
my lease is up. I shall take a new flat somewhere, and on the very first day I
shall have a word with the new housekeeper.
"By the way," I shall say, "about the butter...."
THE WASHING
Of course, it is quite possible to marry for love, but I suspect that a good
many bachelors marry so that they may not have to bother about the
washing any more. That, anyhow, will be one of the reasons with me. "I
offer you," I shall say, "my hand and heart—and the washing; and, oh, do
see that six tablecloths and my footer shorts don't get sent every week."
We affect Hampstead for some reason. Every week a number of shirts

and things goes all the way out to Hampstead and back. I once sent a
Panama to Paris to be cleaned, and for quite a year afterwards I used to lead
the conversation round to travel, and then come out with, "Ah, I well
remember when my Panama was in Paris...." So now, when I am asked at a
dance, "Do you know Hampstead at all?" I reply, "Well, I only know it
slightly myself; but my collars spend about half the year there. They are in
with all the best people."
I can believe that I am not popular in Hampstead, for I give my

laundress a lot of trouble. Take a little thing like handkerchiefs. My rooms,
as I have mentioned, are at the very top of the building, and there is no lift.
Usually I wait till I am just out into the street before I discover that I have
forgotten my handkerchief. It is quite impossible to climb all the stairs
again, so I go and buy one for the day. This happens about three times a
week. The result is that nearly all my handkerchiefs are single ones—there
are no litters of twelve, no twins even, or triplets. Now when you have a lot
of strangers in a drawer like this, with no family ties (or anything) to keep
them together, what wonder if they gradually drift away from each other?
My laundress does her best for them. She works a sort of birthmark in
red cotton in the corner of each, so that she shall know them again. When I
saw it first I was frightened. It looked like the password of some secret
society.
"Are there many aliens in Hampstead?" I asked the housekeeper.
"I don't know, sir."
"Well, look here, what I found on my handkerchief. That's a secret

signal of some sort, you know, that's what it is. I shall get mixed up in some
kind of anarchist row before I know where I am. Will you arrange about
getting my clothes washed somewhere else, please?"
"That's because you haven't got your name on it. She must mark them
somehow."
"Then, why doesn't she mark them with my name? So much simpler."
"It isn't her business to mark your clothes," said the housekeeper.
That, I suppose, is true; but it seems to me that she is giving us both a

lot of unnecessary trouble. Every week I pick out this decorative design
with a penknife, and every week she works it in again. When you consider
the time and the red cotton wasted it becomes clear that a sixpenny bottle of
marking ink and a quill pen would be cheaper to her in the long run.
But she has a weakness for red cotton. The holes in the handkerchiefs
she works round with it—I never quite understand why. To call my
attention to them, perhaps, and to prevent me from falling through. Or else
to say, "You did this. I only washed up to the red, so it can't be my fault."
If I were married and had a house of my own, there would be no man

below; consequently he wouldn't wear the absurd collars he does. I get
about two of them a week (so even red cotton is not infallible); and if they
were the right size and a decent shape I shouldn't grumble so much. But I
do object to my collars mixing in town with these extraordinary things of
his. At Hampstead, it may be, they have to meet on terms of equality, more
or less; force of circumstances throws them together a good deal. But in
town no collar of mine could be expected to keep up the acquaintance. "You
knew me in the bath," I can imagine one of his monstrosities saying; and,
"When I am in the bath I shall know you again," would be the dignified
reply of my "16-Golf."
Collars trouble me a good deal one way or another. Whenever I buy a

new dozen, all the others seem suddenly to have become old-fashioned in
shape, and of the wrong size. Nothing will induce me to wear one of them
again. They get put away in boxes. Covered with dust, they lie forgotten.
Forgotten, did I say? No. The housekeeper finds them and sends them to
the wash. About a month later she finds them again. She is always finding
clothes which have been discarded for ever, and sending them to the wash.
The mistake is, that we have not yet come to an agreement as to what
really is to go to the wash, and what isn't. There is a tacit understanding that
everything on the floor on Monday morning is intended for Hampstead. The
floor is the linen-basket. It seemed a good idea at the time, but it has its
faults. Things gets on to the floor somehow, which were never meant for the
north-west—blankets, and parts of a tweed suit, and sofa cushions. Things
have a mysterious way of dropping. Half-a-dozen pairs of white flannel
trousers dropped from a shelf one December. A pair of footer shorts used to
go every week—a pair which I would carefully put down to take the bath
water when I had finished with it. I wonder what those shorts thought they
were doing. Probably they quite fancied themselves at football, and boasted
about the goals they shot to companions whom they met at Hampstead.
"You're always here!" a pair of local wanderers would say.

"My dear man, I play so hard, I don't care how dirty I get."
The irony of it!
But, worst of all, the laundry-book. Every week the housekeeper says to
me, "Would you pay your book now, as it's been owing for a month?" And
every week I pay. That sounds absurd, but I swear it's true. Or else the
weeks go very quickly.
And such amounts! Great ninepences for a counterpane or a tablecloth

or a white tie. Immense numbers of handkerchiefs, counting (apparently)
twelve as thirteen. Quaint hieroglyphics, which don't mean anything but
seem to get added into the price. And always that little postscript, "As this
has been owing for a month, we must request...."
TAKING STOCK
Beatrice has been spring-cleaning me to-day, or rather my clothes. I

said, wasn't it rather early for it, as none of the birds were singing properly
yet, and she had much better wait till next year; but no, she would do it now.
Beatrice is my sister-in-law, and she said—— Well, I forget what she did
say, but she took a whole bundle of things away with her in a cab; and I
know John will be wearing that purple shirt of mine to-morrow. As a matter
of fact, it was a perfectly new one, and I was only waiting till Lent was
over.
Beatrice said the things were all lying about anyhow, and how I ever
found anything to put on she didn't know; but I could have told her that they
were all arranged on a symmetrical plan of my own. Beatrice doesn't
understand the symmetry of a bachelor's mind. I like a collar in each
drawer, and then, whatever drawer you open, there's a collar ready for you.
Beatrice puts them all in one drawer, and then if you're in a hurry, and open
the wrong drawer by mistake, you probably go up to the office in two
waistcoats and no collar at all. That would be very awkward.
Beatrice actually wanted a braces drawer (if she hadn't married John I
should never let her talk to me about braces), but I explained that I had only
one pair, and was wearing those, so that it would be absurd. I expect she
wanted me to think that John had two pairs. All I can say, is, that, if he has,
he ought to be above taking my best shirt....
I don't think the waistcoat drawer will be a success. There are twenty-
three of them, and some of them don't blend at all well. Twenty-three in one
drawer—you know there are bound to be disputes. I see William has got to
the top already. Ah! he was a fine fellow, the first I ever had. I don't quite
know how to describe him, but in colour he was emerald green, with bits of
red silk peeping through. Sort of open-work, you know, only where you
would expect to see me there was more of William. I wore him at Beatrice's
wedding. He would come. Only he wouldn't let me into the vestry. I wanted
to sign my name; all the others were. I have never worn him since that day,
but Beatrice has fished him out, and now he lies on the very top of the
drawer.
Of course it's awfully good of Beatrice to take so much trouble about

my clothes, and I'm extremely grateful, and after all she did marry my
brother John; but I think sometimes she—— Well, here's a case. You know,
when you have twenty-three waistcoats you perhaps run a bit short of—of
other things. So, naturally, the few you have got left, you—— Well,
Beatrice took them all away, and said that as I couldn't possibly wear them
again she'd cut them up for house-cloths. And really—half-way between
winter and summer is a very awkward time for restocking. But I suppose it
is going to be warmer now?
House-cloths! I bet John has a go at them first.
Beatrice found what they call in the profession a "morning-coat and

vest" under the bed, and said that she would take it away and sell it for me.
I like the way she "finds" things which I have been keeping for years under
the bed. It is absurd to talk about "finding" anything in a small flat, because,
of course, it's there all the time: but Beatrice thought that I ought to be
grateful to her for the discovery, so I pretended I was. She said she would
get at least half-a-crown for it; but I said I would rather have the coat.
However, it turned out that I wasn't even to have had the half-crown....
I used to have thirty pairs of old white gloves in a drawer. I would take
them out sometimes, and stroke them affectionately, and say, "Ah, yes,
those were the ones I wore at that absolutely ripping dance when I first met
Cynthia, and we had supper together. You can see where I spilt the ice
pudding." Or—"This was that Hunt Ball, when I knew nobody and danced
with Hildegarde all the time. She wore black; just look at them now." Well,
Beatrice had that drawer out pretty quick. And now they are on their way to
Perth or Paris, or wherever it is; except Hildegarde's pair, which will just do
for the girl when she cleans the grates. I expect she really will get them you
know; because John doesn't dance.
You know, you mustn't make too much fun of Beatrice; she has ripping
ideas sometimes. She filled a "summer-trunk" for me—a trunk full of all the
clothes I am going to want in the summer. She started with a tennis-racquet
(which, strictly speaking, isn't clothes at all), and went on with some of the
jolliest light waistcoats you ever saw; it made me quite hot to look at them.
Well, now, that's really a good idea so far as it goes. But what will happen
when the summer does come? Why, we shall have to go through the whole
business all over again. And who'll arrange the winter-trunk? Beatrice. And
who'll get the pink pyjamas and the green socks that there's really no room
for, dear? Why, John.
Yet I am sorry for John. He was once as I am. What a life is his now.
Beatrice is a dear, and I will allow no one to say a word against her, but she
doesn't understand that trousers must be folded, not hung; that a collar
which has once been a collar can never be opened out and turned into a cuff
(supposing one wore cuffs); and that a school eleven blazer, even if it
happens to be pink, must not be cut down into a dressing-jacket for the little
one. Poor John! Yes, I am glad now that he has that shirt of mine. It is
perhaps a little too bright for his complexion, perhaps he has not quite the
air to carry it off; but I am glad that it is his. Now I think of it I have a tie
and a pair of socks that would go well with it; and even William—can I part
with William?—yes, he shall have William. Oh, I see that I must be kind to
John.
Dear Beatrice! I wonder when I shall have everything straight again.
MEDES AND PERSIANS
I have already said that I am not afraid of my housekeeper, so there is

no need for me to say it again. There are other motives than fear which
prevent a man from arguing with housekeepers; dislike of conversation with
his intellectual inferiors may be one, the sporting instinct is certainly
another. If one is to play "Medes and Persians" properly, one must be a
sportsman about it. Of course, I could say right out to her "Do this," and she
might do it. Or she could say right out to me, "Do that," and I would reply,
"Don't be absurd." But that wouldn't be the game.
As I play it, a "Mede" is a law which she lays down, and to which (after
many a struggle) in the end I submit; a "Persian" is a law which I lay down,
and to which ... after many a struggle ... in the end ... (when it is too late)....
Well, there are many Medes, but so far I have only scored one Persian of
note.
The first Mede was established last winter. For many weeks I had
opened my bedroom door of a morning to find a small jug of cold water on
the mat outside. The thing puzzled me. What do I want with a small jug of
cold water, I asked myself, when I have quite enough in the bath as it is?
Various happy thoughts occurred to me—as that it was lucky, that it
collected the germs, or (who knows?) indicated a wife with five thousand a
year—but it was a month before the real solution flashed across my mind.
"Perhaps," I said, "it was hot once. But," I added, "it must have been a long
time ago."
The discovery upset me a good deal. In the first place, it is annoying
suddenly to have all one's hopes of a rich wife and protection from disease
dashed to the ground; in the second, I object to anybody but a relation
interfering with my moral character. Here was a comparative stranger trying
to instil the habit of early rising into me by leaving shaving-water outside
the door at three A.M. Was this a thing to be taken lying down?
Decidedly. So I stayed in bed and ignored the water-jug; save that each
morning, as I left my rooms, I gave it a parting sneer. It was gone by the
evening, but turned up again all right next day. After a month I began to get
angry. My housekeeper was defying me; very well, we would see who
could last the longer.
But after two months it was a Mede. Yet I have this triumph over her.
That though I take the water in I ... pour it into the bath and slip back into
bed again. I don't think she knows that.
Since then there have been many Medes. Little ones as to the position of
the chairs; bigger ones as to the number of blankets on the bed. You mustn't
think, though, that I always submit so easily. Sometimes I am firm. In the
matter of "Africa Joe" I have been very firm. Here, I know, I have right on
my side.
A year ago I was presented with a model of an Irish jolting car (with
horse and driver complete), which had been cut out of some sort of black
wood. The thing used to stand over my fireplace. Later on I acquired, at
different times, a grey hippopotamus (in china) and a black elephant. These
I harnessed on in front of the horse; and the whole affair made a very pretty
scene, which was known to my friends as "Sunday in the Forest: Africa Joe
drives his Family to Church." Besides all these I had yet another animal—a
green frog climbing a cardboard ladder. I leant this against the clock. One
had the illusion that the frog was climbing up in order to look at the works
—which was particularly pleasing because the clock didn't go.
Very well. You have the two scenes on the same mantelboard. One, the
frog as Bond Street watchmaker and jeweller, and the other (such is
Empire), Africa Joe in the heart of the forest. And what does the
housekeeper do? If you will believe me, she takes the frog down from the
clock and props him up behind the car, just as though he were getting on to
it in order to go to church with the others!
Now I do put it to you that this is simply spoiling the picture altogether.
Here we have a pleasant domestic episode, such as must occur frequently in
the African forests. Black Joe harnesses his horse, elephant, hippopotamus,
or what not, and drives his family to the eleven o'clock service. And into
this scene of rural simplicity a mere housekeeper elbows her way with
irrevelant frogs and ladders!
It is a mystery to me that she cannot see how absurd her contribution is.
To begin with, the family is in black (save the hippopotamus, who is in a
quiet grey), so is it likely that they would tolerate the presence of a garish
green-and-yellow stranger? (More than likely Joe is a churchwarden, and
has not only himself to think of.) Then, again, consider the title of the
scene: "Africa Joe drives his Family"—not "Africa Joe about to drive." The
horse is trotting, the elephant has one leg uplifted, and even the
hippopotamus is not in a position of rest. How then could the frog put a
ladder up against a moving cart, and climb in? No; here anyhow was a
Mede that must be resisted at all costs. On the question of Africa Joe I
would not be dictated to.
But, after re-emphasising my position daily for three weeks, I saw that
there was only one thing to do. The frog must be sacrificed to the idea of
Empire. So I burnt him.
But it is time I mentioned my one Persian. It was this way. In the winter
I used always to dry myself after the bath in front of my sitting-room fire.
Now, I know all about refraction, and the difficulty of seeing into a room
from outside, and so forth, but this particular room is unusually light,
having six large windows along one of its sides. I thought it proper,
therefore, to draw down the three end blinds by the fireplace; more
especially as the building directly opposite belonged to the Public House
Reform Association. In the fierce light that beats from reform associations
one cannot be too careful.
Little things like blinds easily escape the memory, and it was obvious
that it would be much pleasanter if the housekeeper could be trained always
to leave the end three down. The training followed its usual course.
Every morning I found the blinds up and every morning I drew them
down and left them there. After a month it seemed impossible that I could
ever establish my Persian. But then she forgot somehow; and one day I
woke up to find the three blinds down.
By a real stroke of genius I drew them up as soon as my dressing was

over. Next morning they were down again. I bathed, dried, dressed and
threw them up. She thought it was a Mede, and pulled them down.
But it was a Persian, and, as I pulled them up, I knew that I had scored.
Yet, after all, I am not so sure. For it is now the summer, and I have no
fire, and I do not want the blinds left down. And when I pull them up every
morning I really want to find them up next morning. But I find them down.
So perhaps it really is a Mede. To tell the truth, the distinction between the
two is not so clear as it ought to be. I must try to come to some arrangement
with the housekeeper about it.
THE CUPBOARD
It was the landlord who first called my attention to the cupboard; I

should never have noticed it myself.
"A very useful cupboard you see there," he said, "I should include that
in the fixtures."
"Indeed," said I, not at all surprised; for the idea of his taking away the
cupboard had not occurred to me.
"You won't find many rooms in London with a cupboard like that."

Forecasting and Predictive Analytics With Forecast X 7Th Edition Keating Solutions Manual Full Chapter PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Forecasting and Predictive Analytics With Forecast X 7Th Edition Keating Solutions Manual Full Chapter PDF

Uploaded by

Copyright:

Available Formats

Forecasting and Predictive Analytics

with Forecast X 7th Edition Keating

Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution

Quarter Seasonal Index

Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution

Date Customers MA4 CMA CMAT SF Cf Time

Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution

respectively (numbers are in hundreds). Calculate the MAPE for 2016.

Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution

Propane Gas Sales in Millions of Pounds (Total at End-Month of Each Quarter)

Year March June September December

2005 6.44 4.85 4.67 5.77

2006 6.22 4.25 4.14 5.34

2007 6.07 4.36 4.07 5.84

2008 6.06 4.24 4.20 5.43

2009 6.56 4.25 3.92 5.26

2010 6.65 4.42 4.09 5.51

2011 6.61 4.25 3.98 5.55

2012 6.24 4.34 4.00 5.36

2013 6.40 3.84 3.53 4.74

2014 5.37 3.57 3.32 5.09

2015 6.03 3.98 3.57 4.92

2016 6.16 3.79 3.39 4.51

Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution

Propane Gas Sales (Millions of

Actual, Centered Moving Average and Trend

Actual Forecast Absolute Absolute

Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution

Sales Sales Sales Sales Sales

Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution

Centered Moving Average Trend

Actual and Forecasted Values

c. If actual sales for 2017 were Q1 = 1,445.1, Q2 = 1,683.8, Q3 = 1,586.6, and Q4 =

Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution

Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution

Jet Fuel Sales TSD Model

Actual Forecast Fitted Values

Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution

Jet Fuel Sales: Winters' Model

2. A regression model using just time and quarterly dummy variables.

Jet Fuel Sales: Regression Model

TSD MAPE = 1.34%

Winters' MAPE = 2.67%

Regression MAPE = 5.05%

Upper Midwest Car Sales (CS) (c6p13)

Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution

Upper Midwest Car Sales (000)

Actual Forecast Fitted Values

e. Prepare a Winters’ exponential smoothing forecast of CS using data from 1996Q1

Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution

"Good heavens," I said, "I've only got seven."

There was a shout of laughter.

"Then you'd better begin," said somebody. "Read them out."

I coughed doubtfully, and began;

There was another shout of laughter.

"I am afraid we can't allow that."

"Why ever not?" I asked in amazement.

"Well, you'd hardly find him in a kitchen, would you?"

"Y-y-yes," I said. "Oh y-y-y-yes. Of course."

I pulled myself together with a great effort.

"No, I'm afraid we can't allow it."

"Oh! Oh, well, I suppose that is possible. Go on."

"Gnats," I went on nervously and hastily. "Of course I know that——"