bonjour
je voudrais des améliorations pour ce code sas, graphique de la courbe de roc et aussi l’écriture du modèle avec cette sortie
code :
proc logistic data=memoire.avc_estimation desc;
class Age(ref="3") delais_Ps_Ad(ref="5") cli_Tr_Conscience(ref="1") graviteSigne_Grav_Hemorag_Iv(ref="1")
Type_AVC(ref="1") gravite_Engagement(ref="1") /param=ref;
model pronostic = Age delais_Ps_Ad cli_Tr_Conscience graviteSigne_Grav_Hemorag_Iv
Type_AVC gravite_Engagement;
run;
voici la sortie de sas : la variable explicative est pronostic codée 1=normal; 2=AVCI;3=AVCH

Code : Sélectionner tout - Visualiser dans une fenêtre à part
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
                                        The SAS System     10:23 Thursday, January 15, 2021   1
 
                                    The LOGISTIC Procedure
 
                                      Model Information
 
              Data Set                      MEMOIRE.AVC_ESTIMATION
              Response Variable             Pronostic                  Pronostic
              Number of Response Levels     2
              Model                         binary logit
              Optimization Technique        Fisher's scoring
 
 
                            Number of Observations Read         114
                            Number of Observations Used         114
 
 
                                       Response Profile
 
                              Ordered                       Total
                                Value     Pronostic     Frequency
 
                                    1            2             84
                                    2            1             30
 
                              Probability modeled is Pronostic=2.
 
 
                                   Class Level Information
 
              Class                            Value         Design Variables
 
              Age                              1          1      0
                                               2          0      1
                                               3          0      0
 
              delais_Ps_Ad                     1          1      0      0      0
                                               2          0      1      0      0
                                               3          0      0      1      0
                                               4          0      0      0      1
                                               5          0      0      0      0
 
              cli_Tr_Conscience                1          0
                                               2          1
 
              graviteSigne_Grav_Hemorag_Iv     1          0
                                               2          1
 
              Type_AVC                         1          0      0
                                               2          1      0
                                               3          0      1
 
              gravite_Engagement               1          0

                                        The SAS System     10:23 Thursday, January 15, 2021   2
 
                                    The LOGISTIC Procedure
 
                                   Class Level Information
 
              Class                            Value         Design Variables
 
                                               2          1
 
 
                                   Model Convergence Status
 
                      Quasi-complete separation of data points detected.
 
WARNING: The maximum likelihood estimate may not exist.
WARNING: The LOGISTIC procedure continues in spite of the above warning. Results shown are
         based on the last maximum likelihood iteration. Validity of the model fit is
         questionable.
 
 
                                     Model Fit Statistics
 
                                                         Intercept
                                          Intercept            and
                            Criterion          Only     Covariates
 
                            AIC             133.404         90.014
                            SC              136.140        122.848
                            -2 Log L        131.404         66.014
 
 
                            Testing Global Null Hypothesis: BETA=0
 
                    Test                 Chi-Square       DF     Pr > ChiSq
 
                    Likelihood Ratio        65.3907       11         <.0001
                    Score                   47.2922       11         <.0001
                    Wald                    20.7771       11         0.0358
 
 
                                  Type 3 Analysis of Effects
 
                                                         Wald
                   Effect                    DF    Chi-Square    Pr > ChiSq
 
                   Age                        2        0.4505        0.7983
                   delais_Ps_Ad               4       11.7118        0.0196
                   cli_Tr_Conscience          1        4.6727        0.0306
                   graviteSigne_Grav_He       1        7.8040        0.0052
                   Type_AVC                   2        2.5890        0.2740
                   gravite_Engagement         1        5.4565        0.0195

                                        The SAS System     10:23 Thursday, January 15, 2021   3
 
                                    The LOGISTIC Procedure
WARNING: The validity of the model fit is questionable.
 
                           Analysis of Maximum Likelihood Estimates
 
                                                   Standard          Wald
       Parameter                 DF    Estimate       Error    Chi-Square    Pr > ChiSq
 
       Intercept                  1     13.2538       154.3        0.0074        0.9316
       Age                  1     1     -0.4091      1.1861        0.1190        0.7301
       Age                  2     1     -0.4947      0.7602        0.4234        0.5152
       delais_Ps_Ad         1     1      3.5299      1.4407        6.0033        0.0143
       delais_Ps_Ad         2     1     -0.4385      0.9153        0.2295        0.6319
       delais_Ps_Ad         3     1     -0.9602      1.0582        0.8234        0.3642
       delais_Ps_Ad         4     1     -2.5469      1.4508        3.0820        0.0792
       cli_Tr_Conscience    2     1      1.4780      0.6838        4.6727        0.0306
       graviteSigne_Grav_He 2     1      2.7308      0.9775        7.8040        0.0052
       Type_AVC             2     1    -13.6021       154.3        0.0078        0.9298
       Type_AVC             3     1    -15.1685       154.3        0.0097        0.9217
       gravite_Engagement   2     1      2.0779      0.8895        5.4565        0.0195
 
 
                                     Odds Ratio Estimates
 
                                                  Point          95% Wald
                Effect                         Estimate      Confidence Limits
 
                Age                  1 vs 3       0.664       0.065       6.791
                Age                  2 vs 3       0.610       0.137       2.705
                delais_Ps_Ad         1 vs 5      34.122       2.026     574.593
                delais_Ps_Ad         2 vs 5       0.645       0.107       3.878
                delais_Ps_Ad         3 vs 5       0.383       0.048       3.046
                delais_Ps_Ad         4 vs 5       0.078       0.005       1.345
                cli_Tr_Conscience    2 vs 1       4.384       1.148      16.747
                graviteSigne_Grav_He 2 vs 1      15.345       2.259     104.244
                Type_AVC             2 vs 1      <0.001      <0.001    >999.999
                Type_AVC             3 vs 1      <0.001      <0.001    >999.999
                gravite_Engagement   2 vs 1       7.987       1.397      45.663
 
 
                 Association of Predicted Probabilities and Observed Responses
 
                       Percent Concordant     92.7    Somers' D    0.860
                       Percent Discordant      6.7    Gamma        0.866
                       Percent Tied            0.7    Tau-a        0.336
                       Pairs                  2520    c            0.930

vous m'excuserez de la mauvaise présentation de la discussion, c'est mon premier