Professional Documents
Culture Documents
(2514203202)
Logistics and Supply Chain Management
Demand and Revenue
2nd Assignment/ Sep 23, 2015
Double Exponential Smoothing (Holts) Method application is implemented in the following raw
data:
Mont
h
1
2
3
4
5
6
Mont
h
7
8
9
10
11
12
Dt
502
510
506
530
543
545
Dt
560
550
575
585
580
610
and assuming that the initial series is 505, the data will be analyzed using some combination of and .
Then, the precision of each forecasts will be counted using Mean Absolute Percentage Error (MAPE), Mean
Absolute Deviation (MAD), Mean Squared Error (MSE), and Multi Factor Productivity (MFP).
The function of Holts Model are:
1.
S t Dt (1 )( S t 1 Tt 1 )
Tt ( S t S t 1 ) (1 )Tt 1
2.
3. The equation to generate forecast for the next p period
Ft p S t Tt p
where
St = exponential smoothing value
= the data smoothing constant (0 < < 1)
= the trend estimaton smoothing constant (0 < < 1)
Tt = trend estimation
p = total of the next period which will be forecasted
Month
0
1
0.2
Dt
502
510
506
530
543
545
560
550
575
10
585
11
580
12
610
St
505
506.8
509.74
4
511.30
43
517.24
65
524.85
53
531.70
52
540.45
11
545.83
87
555.23
21
565.14
21
572.46
73
584.47
81
0.1
Tt
0.1
Dt
502
3
2.88
508
2.8864
509.68
2.7537
92
3.0726
3
3.5262
47
3.8586
17
4.3473
39
4.4513
71
4.9455
69
5.4420
16
5.6303
33
6.2683
8
512.63
04
514.05
81
520.31
91
528.38
15
535.56
39
544.79
84
550.29
01
560.17
77
570.58
41
578.09
76
Sum
St
0.1
Tt
MAD
MSE
MAPE
^
Yt Yt
^ | ( Yt Yt
^ ) 2 ( Yt Yt
^ ) /Yt
|Yt Yt
Ft+k
Error
Mont
h
0
1
MFP
-6
36
0.32
0.32
0.1024
6.6304
15.941
89
22.680
88
16.618
46
24.436
15
5.2015
8
24.709
89
24.822
35
9.4158
6
31.902
36
163.41
9
13.618
25
6.6304
43.962
2
254.14
38
514.42
23
276.17
31
597.12
54
27.056
43
610.57
88
616.14
88
88.658
43
1017.7
6
4082.1
32
340.17
77
MFP
15.941
89
22.680
88
16.618
46
24.436
15
5.2015
8
24.709
89
24.822
35
9.4158
6
31.902
36
188.67
98
15.723
32
MAD
0.01195
0.00062
7
-0.0131
0.03007
9
0.04177
0.03049
3
0.04363
6
0.00945
7
0.04297
4
0.04243
1
0.01623
4
0.05229
9
0.28494
5
2.37453
8
MSE
MAPE
Ft+k
^
Yt Yt
^ |
|YtYt
^)
( Yt Yt
505
507.4
3
2.94
508
-6
36
510
510.30
6
2.9366
510.34
-0.34
0.34
0.1156
506
512.51
83
2.8641
74
513.24
26
7.2426
7.2426
52.45525
476
^ ) /Yt
( Yt Yt
0.011952
19
0.000666
67
0.014313
44
530
543
545
560
550
575
10
585
11
580
12
610
516.84
43
522.16
92
527.36
99
533.72
68
538.71
08
545.80
92
553.48
98
560.21
74
569.47
3.0103
49
3.2418
03
3.4376
93
3.7296
18
3.8550
54
4.1793
95
4.5295
09
4.7493
16
5.1996
49
515.38
25
519.85
46
525.41
1
530.80
76
537.45
64
542.56
58
549.98
86
558.01
93
564.96
67
Sum
Error
14.617
49
23.145
39
19.589
05
29.192
45
12.543
59
32.434
17
35.011
36
21.980
72
45.033
33
219.96
49
18.330
41
14.61748
6
23.14538
85
19.58904
69
29.19244
9
12.54358
64
32.43417
42
35.01136
15
21.98071
64
45.03332
87
247.1301
38
20.59417
81
213.6708
97
535.7090
107
383.7307
6
852.1990
805
157.3415
604
1051.975
657
1225.795
433
483.1518
937
2028.000
691
7020.145
838
585.0121
532
0.027580
162
0.042625
025
0.035943
205
0.052129
373
0.022806
521
0.056407
259
0.059848
481
0.037897
787
0.073825
129
0.382130
647
3.184422
054
No.
1.
0.2
0.1
2.
0.1
0.1
3.
0.3
0.1
4.
0.4
0.1
5. *
0.5
0.1
6.
0.1
0.2
7.
0.2
0.2
8.
0.3
0.2
9.
0.4
0.2
10. *
0.5
0.2
11.
0.1
0.3
12.
0.2
0.3
13.
0.5
0.3
14.
0.5
0.4
15.
0.5
0.5
16.
0.5
0.6
17. *
0.5
0.9
18.
0.1
0.5
19.
0.5
0.5
20.
0.6
0.5
21.
0.8
0.5
22.
0.9
0.5
23.
0.6
0.9
MFP
13.618
25
18.330
41
10.471
62
8.3874
2
6.9894
6
17.079
41
11.781
08
8.5019
3
6.5200
5
5.3101
9
15.871
30
10.112
90
4.1074
7
3.2853
8
2.7513
7
2.4227
9
2.0492
8
13.592
17
2.7513
7
2.5086
3
2.2603
0
2.2340
6
1.9051
MAD
15.72332
MSE
340.1777
20.59418
585.0122
12.62091
232.7134
11.66473
184.4816
11.25833
163.0812
19.31381
505.7348
13.85201
269.9642
11.67576
184.5234
11.06867
153.3683
10.71886
143.3416
18.07654
438.2354
12.27309
221.3407
10.32748
137.4608
10.04973
138.2440
9.873685
142.1813
10.05376
147.5819
11.02503
165.7043
15.73967
333.8902
9.87369
142.1813
10.51488
151.5440
11.70105
184.8775
12.21093
11.67443
211.3689
180.9232
MAPE
2.3745
38
3.1844
22
1.8298
19
1.4662
15
1.2203
75
2.9720
56
2.0610
42
1.4919
66
1.1442
83
0.9295
00
2.7667
48
1.7756
91
0.7193
46
0.5740
12
0.4780
50
0.4175
99
0.3444
30
2.3787
81
0.4780
50
0.4300
18
0.3785
13
0.3701
43
0.3160
24. *
0.7
0.9
25.
0.8
0.9
26.
0.9
0.9
27.
0.1
0.9
28.
0.9
0.1
min
1.
0.7
0.9
2.
0.5
0.5
3.
0.5
0.3
4.
0.8
0.9
4
1.8103
3
1.8463
6
1.9956
3
9.6195
3
4.4774
4
1.8103
3
MFP
1.810
33
2.751
37
4.107
47
1.846
36
12.29571
206.7406
12.97398
245.7627
14.81875
300.5574
12.40694
217.4021
11.24948
169.6876
9.873685
137.4608
MAD
12.29
571
9.873
69
10.32
748
12.97
398
MSE
206.7
406
142.1
813
137.4
608
245.7
627
MAPE
0.296
183
0.478
05
0.719
346
0.298
054
50
0.2961
83
0.2980
54
0.3189
88
1.7000
72
0.7707
28
0.2961
83
Error Distribution
300
250
200
150
100
50
0
2
MFP
MAD
MSE
MAPE
MAD
MSE
MAPE
36
48.720
4
30.507
95
6.98
6.98
5.5234
5.5234
22.205
32
5.5345
86
13.953
4
1.9908
3
19.480
1
21.351
07
5.1492
54
17.955
22.205
32
5.5345
86
13.953
42
493.07
63
30.631
64
194.69
8
1.9908
3
19.480
12
3.9634
04
379.47
49
21.351
07
5.1492
54
17.955
32
455.86
83
26.514
81
322.39
36
5
6
7
8
9
10
11
0.0119
5
0.0136
86
0.0109
2
0.0418
97
0.0101
93
0.0256
0.0035
55
0.0354
2
0.0371
32
0.0088
02
0.0309
12
3
21.425
16
21.425
16
459.03
74
6
0.0351
23
MFP
MAD
MSE
MAPE
Mont
h
1
2
3
Dt
502
510
506
530
543
545
560
550
575
10
585
11
580
12
610
St (0.7;
0.9)
503.8
507.906
507.6570
2
523.3384
034
541.3396
243
549.1860
272
559.4027
51
555.8440
35
568.5946
781
583.4552
239
585.3865
965
603.5724
525
St (0.5;
0.5)
505
508.25
508.3125
St (0.5;
0.3)
505
508.55
508.5425
519.7656
25
534.5507
813
545.0556
641
557.7941
895
559.7149
048
570.7465
363
582.3257
179
586.2838
793
601.6919
901
520.1573
75
533.9412
063
543.1919
409
555.5885
171
557.4485
277
569.7612
538
581.7034
288
585.6690
02
601.8014
382
St (0.8;
0.9)
503.2
508.376
507.3804
8
525.3874
304
542.7098
706
547.9832
518
558.8899
867
553.8705
433
570.0798
634
584.8642
258
583.9188
557
604.9082
056
Holt's Model
700
600
Dt
St (0.7; 0.9)
500
St (0.5; 0.5)
400
St (0.5; 0.3)
St (0.8; 0.9)
300
200
100
0
10 11 12
From the tables and graphic above, known that the best forecast model which is suitable with the
existing data is when using initial series 505 and the value of (,) :
a)
b)
c)
d)
(0,7; 0,9)
(0,5; 0,5)
(0,5; 0,3)
(0,8; 0,9)
which the error value are minimum and predict data almost same with the actual.
Mont
h
1
2
3
4
5
6
Mont
h
7
8
9
10
11
12
Dt
502
510
506
530
543
545
Dt
560
550
575
585
580
610
mean
media
n
range
0.2
No.
549.66
67
547.5
502610
0.1
Initial
Series
1.
505
300
400
550
600
0.7
0.9
Initial
Series
No.
1.
505
300
400
550
600
MFP
MAD
MSE
MAPE
13.618
25
60.476
7
37.618
92
3.3322
49
8.0966
4
15.723
32
65.686
51
38.883
02
23.665
14
340.17
77
8335.3
76
2397.7
01
768.25
33
36.199
11
2165.6
54
2.3745
38
11.719
1
7.1607
75
0.3232
94
1.9558
7
MFP
MAD
MSE
MAPE
1.8103
3
1.7816
17
1.7956
24
1.8166
33
1.8236
36
12.295
71
41.418
32
24.351
57
19.487
08
28.869
02
206.74
06
4867.7
16
1325.0
35
543.27
89
1491.2
97
0.2961
83
0.3391
28
0.3181
79
0.2867
56
0.2762
82
From the table above, known that Initial series will be influence the Error, the best initial series is
when the value closer to mean of the data. Because in this case, the precision of forecast measurement
actually using mean to calculate MFP, MAD, MSE, and MAPE.