1. In a time-series decomposition of sales (in millions of units), the following trend has been estimated:
CMAT = 4.7 * 0.37(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.
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 ?gures along with your forecasts to calculate the root-mean-squared error for the forecast period.
2. 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):
|
Mid-Month of Quarter
|
|
Year
|
Feb
|
May
|
Aug
|
Nov
|
Yearly Totals
|
|
2004
|
3.5
|
2.9
|
2.0
|
3.2
|
11.6
|
|
2005
|
4.1
|
3.4
|
2.9
|
3.6
|
14.0
|
|
2006
|
5.2
|
4.5
|
3.1
|
4.5
|
17.3
|
|
2007
|
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 1 through 4.
b. Determine the seasonal index for each quarter.
c. Do the best you can to project the cycle factor through 2008.
d. Make a forecast for each quarter of 2008.
e. The actual numbers of customers served per quarter in 2008 were 6.8, 5.1, 4.7, and 6.5 for quarters 1 through 4, respectively (numbers are in hundreds). Calculate the RMSE for 2008.
f. 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 2004 through 2008 (where data are available).
3. Kim Brite and Larry Short have developed a series of exclusive mobile-home parks in which each unit occupies a site at least 100 ? 150 feet. Each site is well landscaped to provide privacy and a pleasant living environment. Kim and Larry are considering opening more such facilities, but to help manage their cash ?ow they need better forecasts of mobile-home shipments (MHS), since MHS appears to in?uence their vacancy rates and the rate at which they can ?ll newly opened parks. They have 16 years of data on mobile-home shipments, beginning with 1988Q1 and ending with 2003Q4, as shown:
|
Mobile Home Shipments (MHS) (000s)
|
|
Year
|
Q1
|
Q2
|
Q3
|
Q4
|
|
1988
|
56.6
|
49.1
|
58.5
|
57.5
|
|
1989
|
54.9
|
70.1
|
65.8
|
50.2
|
|
1990
|
53.3
|
67.9
|
63.1
|
55.3
|
|
1991
|
63.3
|
81.5
|
81.7
|
69.2
|
|
1992
|
67.8
|
82.7
|
79.0
|
66.2
|
|
1993
|
62.3
|
79.3
|
76.5
|
65.5
|
|
1994
|
58.1
|
66.8
|
63.4
|
56.1
|
|
1995
|
51.9
|
62.8
|
64.7
|
53.5
|
|
1996
|
47.0
|
60.5
|
59.2
|
51.6
|
|
1997
|
48.1
|
55.1
|
50.3
|
44.5
|
|
1998
|
43.3
|
51.7
|
50.5
|
42.6
|
|
1999
|
35.4
|
47.4
|
47.2
|
40.9
|
|
2000
|
43.0
|
52.8
|
57.0
|
57.6
|
|
2001
|
56.4
|
64.3
|
67.1
|
66.4
|
|
2002
|
69.1
|
78.7
|
78.7
|
77.5
|
|
2003
|
79.2
|
86.8
|
87.6
|
86.4
|
Assuming that Kim Brite and Larry Short have hired you as a forecasting consultant:
a. Provide a time-series plot of the actual MHS data along with the deseasonalizeddata. Write a brief memo in which you report the nature and extent of the seasonality in the data. Include seasonal indices in your report.
b. Develop a long-term linear trend for the data, based on the centered moving averages. Let time equal 1 for 1988Q1 in your trend equation. On the basis of this trend, does the future look promising for Brite and Short?
c. One of the things Ms. Brite and Mr. Short are concerned about is the degree towhich MHS is subject to cyclical ?uctuations. Calculate cycle factors and plot themin a time-series graph, including projections of the cycle factor through 2004. In evaluating the cycle factor, see whether interest rates appear to have any effect on the cyclical pattern. The rate for 1988Q1 through 2003Q4 is provided in the following table, should you wish to use this measure of interest rates.
|
Interest Rate
|
|
|
|
|
|
Year
|
Q1
|
Q2
|
Q3
|
Q4
|
|
1988
|
16.4
|
16.3
|
11.6
|
16.7
|
|
1989
|
19.2
|
18.9
|
20.3
|
17.0
|
|
1990
|
16.3
|
16.5
|
14.7
|
12.0
|
|
1991
|
10.9
|
10.5
|
10.8
|
11.0
|
|
1992
|
11.1
|
12.3
|
13.0
|
11.8
|
|
1993
|
10.5
|
10.2
|
9.5
|
9.5
|
|
1994
|
9.4
|
8.6
|
7.9
|
7.5
|
|
1995
|
7.5
|
8.0
|
8.4
|
8.9
|
|
1996
|
8.6
|
8.8
|
9.7
|
10.2
|
|
1997
|
11.0
|
11.4
|
10.7
|
10.5
|
|
1998
|
10.0
|
10.0
|
10.0
|
10.0
|
|
1999
|
9.2
|
8.7
|
8.4
|
7.6
|
|
2000
|
6.5
|
6.5
|
6.0
|
6.0
|
|
2001
|
6.0
|
6.0
|
6.0
|
6.0
|
|
2002
|
6.0
|
6.9
|
7.5
|
8.1
|
|
2003
|
8.8
|
9.0
|
8.8
|
8.7
|
d. Demonstrate for Ms. Brite and Mr. Short how well your time-series decomposition model follows the historical pattern in the data by plotting the actual values of MHS and those estimated by the model in a single time-series plot.
e. Prepare a forecast for 2004 and calculate the root-mean-squared error (RMSE), given the actual values of MHS for 2004 shown:
|
MHS
|
|
Period Forecast
|
Actual Squared Error
|
|
2004Q1
|
35.4
|
|
2004Q2
|
47.3
|
|
2004Q3
|
47.2
|
|
2004Q4
|
40.9
|
Sum of squared errors =
Mean-squared error =
Root-mean-squared error =
4. a. Use the following data on millions of dollars of jewelry sales (JS) to prepare a time-series decomposition forecast of JS for the four quarters of 2005:
|
Date
|
Jewelry Sales
($ Millions)
|
Date
|
Jewelry Sales
($ Millions)
|
Date
|
Jewelry Sales
($ Millions)
|
|
Jan-94
|
904
|
May-94
|
1,367
|
Sep-94
|
1,246
|
|
Feb-94
|
1,191
|
Jun-94
|
1,257
|
Oct-94
|
1,323
|
|
Mar-94
|
1,058
|
Jul-94
|
1,224
|
Nov-94
|
1,731
|
|
Apr-94
|
1,171
|
Aug-94
|
1,320
|
Dec-94
|
4,204
|
|
Jan-95
|
914
|
Sep-98
|
1,372
|
May-02
|
2,120
|
|
Feb-95
|
1,223
|
Oct-98
|
1,506
|
Jun-02
|
1,667
|
|
Mar-95
|
1,138
|
Nov-98
|
1,923
|
Jul-02
|
1,554
|
|
Apr-95
|
1,204
|
Dec-98
|
5,233
|
Aug-02
|
1,746
|
|
May-95
|
1,603
|
Jan-99
|
1,163
|
Sep-02
|
1,503
|
|
Jun-95
|
1,388
|
Feb-99
|
1,662
|
Oct-02
|
1,662
|
|
Jul-95
|
1,259
|
Mar-99
|
1,402
|
Nov-02
|
2,208
|
|
Aug-95
|
1,393
|
Apr-99
|
1,468
|
Dec-02
|
5,810
|
|
Sep-95
|
1,325
|
May-99
|
1,877
|
Jan-03
|
1,361
|
|
Oct-95
|
1,371
|
Jun-99
|
1,635
|
Feb-03
|
2,019
|
|
Nov-95
|
1,867
|
Jul-99
|
1,596
|
Mar-03
|
1,477
|
|
Dec-95
|
4,467
|
Aug-99
|
1,617
|
Apr-03
|
1,616
|
|
Jan-96
|
1,043
|
Sep-99
|
1,530
|
May-03
|
2,071
|
|
Feb-96
|
1,439
|
Oct-99
|
1,653
|
Jun-03
|
1,711
|
|
Mar-96
|
1,316
|
Nov-99
|
2,179
|
Jul-03
|
1,677
|
|
Apr-96
|
1,359
|
Dec-99
|
6,075
|
Aug-03
|
1,761
|
|
May-96
|
1,768
|
Jan-00
|
1,253
|
Sep-03
|
1,629
|
|
Jun-96
|
1,408
|
Feb-00
|
1,991
|
Oct-03
|
1,759
|
|
Jul-96
|
1,375
|
Mar-00
|
1,510
|
Nov-03
|
2,291
|
|
Aug-96
|
1,477
|
Apr-00
|
1,570
|
Dec-03
|
6,171
|
|
Sep-96
|
1,332
|
May-00
|
2,139
|
Jan-04
|
1,461
|
|
Oct-96
|
1,462
|
Jun-00
|
1,783
|
Feb-04
|
2,344
|
|
Nov-96
|
1,843
|
Jul-00
|
1,643
|
Mar-04
|
1,764
|
|
Dec-96
|
4,495
|
Aug-00
|
1,770
|
Apr-04
|
1,826
|
|
Jan-97
|
1,041
|
Sep-DO
|
1,705
|
May-04
|
2,226
|
|
Feb-97
|
1,411
|
Oct-00
|
1,681
|
Jun-04
|
1,882
|
|
Mar-97
|
1,183
|
Nov-00
|
2,174
|
Jul-04
|
1,787
|
|
Apr-97
|
1,267
|
Dec-00
|
5,769
|
Aug-04
|
1,794
|
|
May-97
|
1,597
|
Jan-01
|
1,331
|
Sep-04
|
1,726
|
|
Jun-97
|
1,341
|
Feb-01
|
1,973
|
Oct-04
|
1,845
|
|
Jul-97
|
1,322
|
Mar-01
|
1,580
|
Nov-04
|
2,399
|
|
Aug-97
|
1,359
|
Apr-01
|
1,545
|
Dec-04
|
6,489
|
|
Sep-97
|
1,344
|
May-01
|
1,992
|
Jan-05
|
?
|
|
Oct-97
|
1,406
|
Jun-01
|
1,629
|
Feb-05
|
?
|
|
Nov-97
|
1,813
|
Jul-01
|
1,530
|
Mar-05
|
?
|
|
Dec-97
|
4,694
|
Aug-01
|
1,679
|
Apr-05
|
?
|
|
Jan-98
|
1,119
|
Sep-01
|
1,394
|
May-05
|
?
|
|
Feb-98
|
1,513
|
Oct-01
|
1 586
|
Jun-05
|
?
|
|
Mar-98
|
1,238
|
Nov-01
|
2,152
|
Jul-05
|
?
|
|
Apr-98
|
1,362
|
Dec-01
|
5,337
|
Aug-05
|
|
|
May-98
|
1,756
|
Jan-02
|
1,304
|
Sep-05
|
7
|
|
Jun-98
|
1,527
|
Feb-02
|
2,004
|
Oct-05
|
|
|
Jul-98
|
1,415
|
Mar-02
|
1,612
|
Nov-05
|
|
|
Aug-98
|
1,466
|
Apr-02
|
1,626
|
Dec-05
|
|
The actual data for 2005 are:
|
Date
|
Jewelry Sales (S Millions)
|
|
Jan-OS
|
1,458
|
|
Feb-OS
|
2,394
|
|
Mar-OS
|
1,773
|
|
Apr-OS
|
1,909
|
|
May-05
|
2,243
|
|
Jun-05
|
1,953
|
|
Jul-05
|
1,754
|
|
Aug-05
|
1,940
|
|
Sep-OS
|
1,743
|
|
Oct-05
|
1,878
|
|
Nov-OS
|
2,454
|
|
Dec-OS
|
6,717
|
b. Evaluate your model in terms of ?t and accuracy using RMSE.
c. Plot your forecast values of JS along with the actual values.
d. Look at the seasonal indices, and explain why you think they do or do not make sense.
e. Compare the results from your time-series decomposition model with those obtained using a Winters' exponential smoothing model in terms of both ?t and accuracy.
5. Estimating the volume of loans that will be made at a credit union is crucial to effective cash management in those institutions. In the table that follows are quarterly data for a real credit union located in a midwestern city. Credit unions are ?nancial institutions similar to banks, but credit unions are not-for-pro?t ?rms whose members are the actual owners (remember their slogan, "It's where you belong"). The members may be both depositors in and borrowers from the credit union.
|
Quarter
|
Loan Volume
|
Assets
|
Members
|
Prime Rate
|
|
Mar-98
|
2,583,780
|
4,036,810
|
3,522
|
6.25
|
|
Jun-98
|
2,801,100
|
4,164,720
|
3,589
|
6.75
|
|
Sep-98
|
2,998,240
|
4,362,680
|
3,632
|
7.13
|
|
Dec-98
|
3,032,720
|
4,482,990
|
3,676
|
7.75
|
|
Mar-99
|
3,094,580
|
4,611,300
|
3,668
|
8
|
|
Jun-99
|
3,372,680
|
4,696,720
|
3,689
|
8.63
|
|
Sep-99
|
3,499,350
|
4,844,960
|
3,705
|
9.41
|
|
Dec-99
|
3,553,710
|
4,893,450
|
3,722
|
11.55
|
|
Mar-00
|
3,651,870
|
5,089,840
|
3,732
|
11.75
|
|
Jun-00
|
3,832,440
|
5,185,360
|
3,770
|
11.65
|
|
Sep-00
|
4,013,310
|
5,381,140
|
3,845
|
12.9
|
|
Dec-00
|
3,950,100
|
5,413,720
|
3,881
|
15.3
|
|
Mar-01
|
3,925,100
|
5,574,160
|
3,923
|
18.31
|
|
Jun-01
|
3,717,4,80
|
5,838,990
|
3,941
|
12.63
|
|
Sep-01
|
3,712,300
|
6,150,350
|
3,955
|
12.23
|
|
Dec-01
|
3,677,940
|
6,133,030
|
3,943
|
20.35
|
|
Mar-02
|
3,724,770
|
6,119,030
|
3,960
|
18.05
|
|
Jun-02
|
3,787,760
|
6,221,090
|
3,971
|
20.03
|
|
Sep-02
|
3,981,620
|
6,229,000
|
3,993
|
20.08
|
|
Dec-02
|
3,848,660
|
6,412,230
|
4,011
|
15.75
|
|
Mar-03
|
3,619,830
|
6,795,830
|
4,040
|
16.5
|
|
Jun-03
|
3,623,590
|
7,538,210
|
4,103
|
16.5
|
|
Sep-03
|
3,632,120
|
8,496,080
|
4,133
|
13.5
|
|
Dec-03
|
3,482,000
|
9,979,390
|
4,173
|
11.5
|
|
Mar-04
|
3,378,500
|
11,475,300
|
4,218
|
10.5
|
|
Jun-04
|
3,433,470
|
12,116,900
|
4,266
|
10.5
|
|
Sep-04
|
3,615,430
|
12,686,500
|
4,305
|
11
|
|
Dec-04
|
3,865,780
|
13,457,600
|
4,657
|
11
|
|
Mar-OS
|
3,955,270
|
14,118,300
|
4,741
|
11.21
|
|
Jun-05
|
4,394,140
|
14,448,600
|
4,826
|
12.6
|
|
Sep-OS
|
4,803,630
|
14,687,200
|
4,943
|
12.97
|
|
Dec-05
|
4,952,740
|
14,885,800
|
4,945
|
11.06
|
|
Mar-06
|
5,249,760
|
16,106,300
|
5,007
|
10.5
|
|
Jun-06
|
5,943,390
|
17,079,400
|
5,112
|
9.78
|
|
Sep-06
|
6,387,000
|
17,846,800
|
5,164
|
9.5
|
|
Dec-06
|
6,435,750
|
19,435,600
|
5,210
|
9.5
|
|
Mar-07
|
6,482,780
|
19,714,100
|
5,255
|
9.1
|
|
Jun-07
|
6,683,800
|
21,185,800
|
5,289
|
8.5
|
|
Sep-07
|
7,094,210
|
22,716,700
|
5,391
|
7.5
|
|
Dec-07
|
7,329,770
|
23,790,500
|
5,461
|
7.5
|
a. Estimate a multiple-regression model to estimate loan demand and calculate its root mean squared error.
b. Estimate a time-series decomposition model to estimate loan demand with the same data and calculate its root-mean-squared error.
c. Combine the models in parts (a) and (b) and determine whether the combined model performs better than either or both of the original models. Try to explain why you obtained the results you did.
6. HeathCo Industries, a producer of a line of skiwear, has been the subject of exercises in several earlier chapters of the text. The data for its sales and two potential causal variables, income (INCOME) and the northern-region unemployment rate (NRUR), are repeated in the following table:
|
Date
|
Sales
|
Income
|
NRUR
|
|
|
Jan-98
|
72,962
|
218
|
8.4
|
|
|
Apr-98
|
81,921
|
237
|
8.2
|
|
|
Jul-98
|
97,729
|
263
|
8.4
|
|
|
Oct-98
|
142,161
|
293
|
8.4
|
|
|
Jan-99
|
145,592
|
318
|
8.1
|
|
|
Apr-99
|
117,129
|
359
|
7.7
|
|
|
Jul-99
|
114,159
|
404
|
7.5
|
|
|
Oct-99
|
151,402
|
436
|
7.2
|
|
|
Jan-00
|
153,907
|
475
|
6.9
|
|
|
Apr-00
|
100,144
|
534
|
6.5
|
|
|
Jul-00
|
123,242
|
574
|
6.5
|
|
|
Oct-00
|
128,497
|
622
|
6.4
|
|
|
Jan-01
|
176,076
|
667
|
6.3
|
|
|
Apr-01
|
180,440
|
702
|
6.2
|
|
|
Jul-01
|
162,665
|
753
|
6.3
|
|
|
Oct-01
|
220,818
|
796
|
6.5
|
|
|
Jan-02
|
202,415
|
858
|
6.8
|
|
|
Apr-02
|
211,780
|
870
|
7.9
|
|
|
Jul-02
|
163,710
|
934
|
8.3
|
|
|
Oct-02
|
200,135
|
1,010
|
8
|
|
|
Jan-03
|
174,200
|
1,066
|
8
|
|
|
Apr-03
|
182,556
|
1,096
|
8
|
|
|
Jul-03
|
198,990
|
1,162
|
8
|
|
|
Oct-03
|
243,700
|
1,178
|
8.9
|
|
|
Jan-04
|
253,142
|
1,207
|
9.6
|
|
|
Apr-04
|
218,755
|
1,242
|
10.2
|
|
|
Jul-04
|
225,422
|
1,279
|
10.7
|
|
|
|
|
|
|
|
|
Date
|
Sales
|
Income
|
NRUR
|
|
Oct-04
|
253,653
|
1,318
|
11.5
|
|
|
Jan-05
|
257,156
|
1,346
|
11.2
|
|
|
Apr-OS
|
202,568
|
1,395
|
11
|
|
|
Jul-05
|
224,482
|
1,443
|
10.1
|
|
|
Oct-05
|
229,879
|
1,528
|
9.2
|
|
|
Jan-06
|
289,321
|
1,613
|
8.5
|
|
|
Apr-06
|
266,095
|
1,646
|
8
|
|
|
Jul-06
|
262,938
|
1,694
|
8
|
|
|
Oct-06
|
322,052
|
1,730
|
7.9
|
|
|
Jan-07
|
313,769
|
1,755
|
7.9
|
|
|
Apr-07
|
315,011
|
1,842
|
7.9
|
|
|
Jul-07
|
264,939
|
1,832
|
7.8
|
|
|
Oct-07
|
301,479
|
1,882
|
7.6
|
|
|
Jan-08
|
334,271
|
1,928
|
7.6
|
|
|
Apr-08
|
328,982
|
1,972
|
7.7
|
4- Holdout
|
|
Jul-08
|
317,921
|
2,017
|
7.5
|
|
|
Oct-08
|
350,118
|
2,062
|
7.4
|
|
|
|
|
|
|
|
|
|
|
a. Develop a multiple-regression model of SALES as a function of both INCOME and NRUR: SALES = a + b1(INCOME) + b2(NRUR)
Use this model to forecast sales for 2008Q1-2008Q4 (call your regression forecast series SFR), given that INCOME and NRUR for 2004 have been forecast to be:
|
Quarter
|
INCOME
|
NRUR
|
|
2008Q1
|
1,928
|
7.6
|
|
2008Q2
|
1,972
|
7.7
|
|
2008Q3
|
2,017
|
7.5
|
|
2008Q4
|
2,062
|
7.4
|
b. Calculate the RMSE for your regression model for both the historical period (1998Q1-2007Q4) and the forecast horizon (2008Q1-2008Q4).
|
Period
|
RMSE
|
|
Historical
|
|
|
Forecast
|
|
c. Now prepare a forecast through the historical period and the forecast horizon (2008Q1-2008Q4) using Winters' exponential smoothing. Call this forecast series SFW, and ?ll in the RMSEs for SFW:
|
Period
|
RMSE
|
|
Historical
|
|
|
Forecast
|
|
d. Solely on the basis of the historical data, which model appears to be the best? Why?
e. Now prepare a combined forecast (SCF) using the regression technique described in this chapter. In the standard regression:
SALES = a + h1(SFR) + /22(SFW)
Is the intercept essentially zero? Why? If it is, do the following regression as a basis for developing SCF:
SALES = b1(SFR) + h2(SFW)
f. Calculate the RMSEs for SCF:
|
Period
|
RMSE
|
|
Historical
|
|
|
Forecast
|
|
Did combining models reduce the RMSE in the historical period? What about the actual forecast?
7. Your company produces a favorite summertime food product, and you have been placed in charge of forecasting shipments of this product. The historical data below represent your company's past experience with the product.
a. Since the data appear to have both seasonality and trend, you should estimate a Winters' model and calculate its root-mean-squared error.
b. You also have access to a survey of the potential purchasers of your product. This information has been collected for some time, and it has proved to be quite accurate for predicting shipments in the past. Calculate the root-mean-squared error of the purchasers' survey data.
c. After checking for bias, combine the forecasts in parts (a) and (b) and determine if a combined model may forecast better than either single model.
|
Date
|
Shipments ($000)
|
Purchasers'
Survey ($000)
|
Date
|
Shipments ($000)
|
Purchasers'
Survey ($000)
|
|
Apr-2002
|
13,838.00
|
13,920.32
|
Jun-2003
|
21,056.00
|
24,644.20
|
|
May-2002
|
15,137.00
|
15,052.82
|
Jul-2003
|
13,509.00
|
14,224.17
|
|
Jun-2002
|
23,713.00
|
26,207.69
|
Aug-2003
|
9,729.00
|
9,194.77
|
|
Jul-2002
|
17,141.00
|
17,237.59
|
Sep-2003
|
13,454.00
|
12,141.25
|
|
Aug-2002
|
7,107.00
|
7,687.23
|
Oct-2003
|
13,426.00
|
11,971.93
|
|
Sep-2002
|
9,225.00
|
9,788.06
|
Nov-2003
|
17,792.00
|
17,654.14
|
|
Oct-2002
|
10,950.00
|
7,889.46
|
Dec-2003
|
19,026.00
|
15,580.19
|
|
Nov-2002
|
14,752.00
|
14,679.10
|
Jan-2004
|
9,432.00
|
9,961.98
|
|
Dec-2002
|
18,871.00
|
17,644.48
|
Feb-2004
|
6,356.00
|
7,368.55
|
|
Jan-2003
|
11,329.00
|
10,436.45
|
Mar-2004
|
12,893.00
|
11,286.25
|
|
Feb-2003
|
6,555.00
|
6,304.89
|
Apr-2004
|
19,379.00
|
18,915.33
|
|
Mar-2003
|
9,335.00
|
9,354.44
|
May-2004
|
14,542.00
|
14,056.06
|
|
Apr-2003
|
10,845.00
|
11,759.15
|
Jun-2004
|
18,043.00
|
20,699.38
|
|
May-2003
|
15,185.00
|
14,971.57
|
Jul-2004
|
10,803.00
|
12,892.97
|