数据模型与决策作业.docx
《数据模型与决策作业.docx》由会员分享,可在线阅读,更多相关《数据模型与决策作业.docx(31页珍藏版)》请在冰豆网上搜索。
数据模型与决策作业
数据模型与决策作业
一、Cropain公司基建部问题
1、首先将60组数据单独列出,找到因变量Y(earn)和自变量X(size、p15、inc、nrest、price),数据如下所示:
STOR
EARN
SIZE
P15
INC
NREST
PRICE
1
28.3
129
980
27.6
45
16.10
2
-1.5
91
1290
28.3
27
11.40
3
68.9
140
2940
30.2
5
21.70
4
202.1
184
3570
27.6
7
11.80
5
115.8
144
1700
33.9
25
16.60
6
221.7
160
4640
32.5
8
22.10
7
292.9
94
3600
33.1
89
24.30
8
134.4
100
3450
29.7
14
16.40
9
37.4
85
1930
28.4
43
12.90
10
181.0
92
3520
28.3
76
13.00
11
246.9
167
3970
38.3
9
22.80
12
178.3
199
3190
32.1
11
10.10
13
214.9
83
4920
36.0
8
16.70
14
0.6
141
1210
35.3
11
31.00
15
252.3
240
4150
35.9
16
14.00
16
124.2
82
2790
33.4
6
13.30
17
258.1
96
4180
35.4
9
12.30
18
193.0
78
4650
29.7
51
28.80
19
54.8
99
1320
28.8
17
12.80
20
45.4
75
2210
29.8
13
21.70
21
66.3
52
4180
28.6
26
15.90
22
123.7
177
1450
34.9
14
12.90
23
57.8
111
1670
26.4
41
11.70
24
75.2
84
720
28.0
64
16.70
25
115.6
95
2620
28.2
50
10.90
26
140.9
95
3990
29.6
21
23.20
27
94.1
67
3490
26.9
18
16.00
28
250.8
154
3950
27.7
64
14.30
29
-43.0
93
2570
25.2
5
33.40
30
145.6
67
1990
36.5
61
11.10
31
147.6
61
4790
32.8
31
7.70
32
175.1
116
4460
27.7
19
12.90
33
117.9
92
2050
33.6
5
11.20
34
79.5
78
3370
29.5
45
20.00
35
140.5
103
2450
34.9
13
32.90
36
399.2
191
4850
34.5
96
18.20
37
246.3
263
2880
38.8
29
22.60
38
77.6
261
770
31.0
13
20.60
39
108.3
169
1770
31.8
10
19.30
40
188.6
97
4330
30.8
29
10.50
41
143.5
117
3310
29.9
36
19.50
42
175.5
116
1550
34.0
44
12.50
43
94.1
76
4050
31.0
19
11.70
44
214.2
144
3920
30.0
26
7.60
45
63.3
87
3230
25.5
32
18.70
46
237.1
73
5150
35.2
14
10.50
47
208.8
59
3450
34.4
71
11.70
48
110.6
83
4070
29.5
44
25.00
49
165.4
125
2800
33.8
12
11.30
50
-11.4
56
2150
29.9
12
14.10
51
216.3
146
2800
32.1
26
11.60
52
65.7
62
2020
32.7
70
18.00
53
67.6
96
2320
30.0
7
13.60
54
127.9
86
2480
34.4
17
16.50
55
82.9
88
1870
28.8
16
12.80
56
-2.9
72
3310
28.7
10
24.30
57
247.7
119
3620
33.4
63
13.30
58
343.0
285
4160
27.6
40
18.30
59
193.1
193
1950
28.7
34
12.50
60
277.5
92
4890
36.0
31
14.10
打开EXCEL表格-工具-数据分析-回归-确定-Y值区域为EARN列,X值区域为SIZE到PRICE列,点标志-确定,生成数据如下:
SUMMARYOUTPUT
回归统计
MultipleR
0.92532
RSquare
0.856217
AdjustedRSquare
0.842903
标准误差
36.20023
观测值
60
方差分析
df
SS
MS
F
SignificanceF
回归分析
5
421397.1
84279.41
64.313
1.64E-21
残差
54
70764.67
1310.457
总计
59
492161.7
Coefficients
标准误差
tStat
P-value
Lower95%
Upper95%
下限95.0%
上限95.0%
Intercept
-353.82
48.33297
-7.32047
1.24E-09
-450.722
-256.918
-450.722
-256.918
SIZE
0.771877
0.089646
8.6103
1.03E-11
0.592148
0.951606
0.592148
0.951606
P15
0.044175
0.004035
10.94786
2.52E-15
0.036086
0.052265
0.036086
0.052265
INC
8.776491
1.485928
5.906402
2.41E-07
5.797384
11.7556
5.797384
11.7556
NREST
1.412648
0.210493
6.711147
1.21E-08
0.990635
1.834661
0.990635
1.834661
PRICE
-2.68531
0.796325
-3.37213
0.001385
-4.28185
-1.08878
-4.28185
-1.08878
回归方程如下:
Y(earn)=0.77size+0.04p15+8.78inc+1.41nrest-2.69price-353.82
多重共线性检验如下:
回到EXCEL表格-工具-数据分析-相关系数-确定,选定区域为从EARN到PRICE的所有列,点击标志在第一行,确定,生成相关性系数.
EARN
SIZE
P15
INC
NREST
PRICE
EARN
1
SIZE
0.436623
1
P15
0.62823
-0.05286
1
INC
0.464943
0.180274
0.154923
1
NREST
0.337582
-0.09639
0.067913
-0.05825
1
PRICE
-0.18002
0.066436
-0.02547
0.004017
-0.06345
1
2、将Y变量(earn)和X变量(从size到price)粘贴到MINITAB中,统计-回归-逐步回归-响应(earn),预测变量(从size到price)-确定,得出数据如下:
逐步回归:
EARN与SIZE,P15,INC,NREST,PRICE
入选用Alpha:
0.15删除用Alpha:
0.15
响应为5个自变量上的EARN,N=50
步骤12345
常量-0.6348-354.7460-421.2498-412.5582-379.4336
P150.04590.04160.03970.04360.0432
T值5.816.357.2010.0010.49
P值0.0000.0000.0000.0000.000
INC11.712.89.79.7
T值4.936.375.856.19
P值0.0000.0000.0000.000
NREST1.331.481.46
T值4.566.436.70
P值0.0000.0000.000
SIZE0.610.63
T值5.516.04
P值0.0000.000
PRICE-2.01
T值-2.55
P值0.014
S68.055.846.836.534.5
R-Sq41.2661.2873.3384.0986.14
R-Sq(调整)40.0459.6371.5982.6784.56
MallowsCp140.478.942.610.56.0
五个变量的回归方程如下:
Earn=0.0432p15+9.7inc+1.46nrest+0.63size-2.01price-379.43;
数据中51到60相关数据如下:
STOR
EARN
K
SIZE
P15
INC
NREST
PRICE
51
216.3
776
146
2800
32.1
26
11.60
52
65.7
648
62
2020
32.7
70
18.00
53
67.6
690
96
2320
30.0
7
13.60
54
127.9
715
86
2480
34.4
17
16.50
55
82.9
650
88
1870
28.8
16
12.80
56
-2.9
788
72
3310
28.7
10
24.30
57
247.7
782
119
3620
33.4
63
13.30
58
343.0
1558
285
4160
27.6
40
18.30
59
193.1
936
193
1950
28.7
34
12.50
60
277.5
688
92
4890
36.0
31
14.10
各组实际利润率=earn/k,因而51到60的每组实际利润率如下:
STOR
EARN
K
SIZE
P15
INC
NREST
PRICE
实际利润率
51
216.3
776
146
2800
32.1
26
11.60
0.278719
52
65.7
648
62
2020
32.7
70
18.00
0.101418
53
67.6
690
96
2320
30.0
7
13.60
0.098048
54
127.9
715
86
2480
34.4
17
16.50
0.178936
55
82.9
650
88
1870
28.8
16
12.80
0.127586
56
-2.9
788
72
3310
28.7
10
24.30
-0.00365
57
247.7
782
119
3620
33.4
63
13.30
0.316819
58
343.0
1558
285
4160
27.6
40
18.30
0.220187
59
193.1
936
193
1950
28.7
34
12.50
0.206441
60
277.5
688
92
4890
36.0
31
14.10
0.403396
各组的预测利润=变量回归方程上各项*相关各项的数据
举例:
第51店的预测利润为Earn=0.0432*2800+9.7*32.1+1.46*26+0.63*146-2.0111.6-379.43=159.524;
因而各组利润为:
STOR
EARN
K
SIZE
P15
INC
NREST
PRICE
预测利润
51
216.3
776
146
2800
32.1
26
11.60
159.524
52
65.7
648
62
2020
32.7
70
18.00
130.104
53
67.6
690
96
2320
30.0
7
13.60
55.158
54
127.9
715
86
2480
34.4
17
16.50
107.221
55
82.9
650
88
1870
28.8
16
12.80
33.786
56
-2.9
788
72
3310
28.7
10
24.30
53.069
57
247.7
782
119
3620
33.4
63
13.30
241.151
58
343.0
1558
285
4160
27.6
40
18.30
269.169
59
193.1
936
193
1950
28.7
34
12.50
129.305
60
277.5
688
92
4890
36.0
31
14.10
255.897
各组预测利润率、与实际利润率比较为:
STOR
EARN
K
SIZE
P15
INC
NREST
PRICE
预测利润
预测利润率
实际利润率
51
216.3
776
146
2800
32.1
26
11.60
159.524
20.56%
27.87%
52
65.7
648
62
2020
32.7
70
18.00
130.104
20.09%
10.14%
53
67.6
690
96
2320
30.0
7
13.60
55.158
8.00%
9.80%
54
127.9
715
86
2480
34.4
17
16.50
107.221
15.00%
17.89%
55
82.9
650
88
1870
28.8
16
12.80
33.786
5.20%
12.76%
56
-2.9
788
72
3310
28.7
10
24.30
53.069
6.73%
-0.37%
57
247.7
782
119
3620
33.4
63
13.30
241.151
30.84%
31.68%
58
343.0
1558
285
4160
27.6
40
18.30
269.169
17.28%
22.02%
59
193.1
936
193
1950
28.7
34
12.50
129.305
13.82%
20.64%
60
277.5
688
92
4890
36.0
31
14.10
255.897
37.20%
40.34%
60组数据的相关系数和回归方程如下所示:
逐步回归:
EARN与SIZE,P15,INC,NREST,PRICE
入选用Alpha:
0.15删除用Alpha:
0.15
响应为5个自变量上的EARN,N=60
步骤12345
常量-3.083-103.061-145.274-399.009-353.820
P150.04820.05010.04850.04440.0442
T值6.157.959.1010.1010.95
P值0.0000.0000.0000.0000.000
SIZE0.7980.8520.7540.772
T值5.737.237.738.61
P值0.0000.0000.0000.000
NREST1.391.451.41
T值4.936.346.71
P值0.0000.0000.000
INC8.88.8
T值5.445.91
P值0.0000.000
PRICE-2.69
T值-3.37
P值0.001
S71.757.648.539.536.2
R-Sq39.4761.6073.2282.5985.62
R-Sq(调整)38.4260.2671.7981.3384.29
MallowsCp171.390.248.615.46.0
Y(earn)=0.0442p15+0.772size+1.41nrest+8.8inc-2.69price-353.820
根据60组数据的预测回归方程对未来10组数据进行预测如下:
STOR
EARN
K
SIZE
P15
INC
NREST
PRICE
预测利润
预测利润率
Calais
660
54
600
38
18
22
19.315
2.93%
Montchanin
733
120
1300
31
21
13
67.705
9.24%
Aubusson
1050
135
2210
29
13
22
63.239
6.02%
Toulouse
836
245
3400
37
62
13
364.995
43.66%
Torcy
784
96
260
30
38
18
0.944
0.12%
Marseilles-1
925
197
1650
23
41
12
99.124
10.72%
Marseilles-2
1090
93
2570
25
5
33
-29.466
-2.70%
Clermont
738
169
780
30
11
9
67.304
9.12%
Montpellier
584
149
2500
29
26
13
124.002
21.23%
Dijon
681
150
1650
35
54
15
176.744
25.95%
二、菲拉托伊·里尤尼蒂纺织厂问题
首先,把要求的相关决策变量清空,如下所示:
DECISIONVARIABLES
Productboughtfromeachsupplier(Kg/month)
Supplier
Size
Extrafine
Fine
Medium
Coarse
Ambrosi
Bresciani
Castri
DeBlasi
Estensi
FilatoiR.
Giuliani
根据题意,本题给出了各工厂生产机器的单位时间和月度最大使用时间,和各工厂生产四种产品的单位成本和运输成本,要求的是在满足各品种需求量的基础上各工厂如何生产总成本最低的问题,根据题意,首先建立目标函数:
COSTOFPRODUCTION
($/Kg)
Supplier
Size
Extrafine
Fine
Medium
Coarse
Ambrosi
13.00
10.65
9.60
Bresciani
17.40
14.10
11.20
9.45
Castri
17.40
14.22
11.00
9.50
DeBlasi
14.30
11.25
9.60
Estensi
17.50
13.80
11.40
9.60
FilatoiR.
18.25
13.90
11.40
8.90
Giuliani
19.75
13.90
10.75
9.40
COSTOFTRANSPORTATION
($/Kg)
Supplier
Size
Extrafine
Fine
Medium
Coarse
Ambrosi
0.30
0.30
0.45
0.45
Bresciani
0.40
0.40
0.60
0.60
Castri
0.80
0.80
1.20
1.20
DeBlasi
0.70
0.70
1.05
1.05
Estensi
0.70
0.70
1.05
1.05
FilatoiR.
-
-
-
-
Giuliani
0.50
0.50
0.75
0.75