我用stata想进行分层模型分析,用的melogit命令,环境变量是一个连续变量,分析结果如下,怎么看出环境变量的OR值和显著性?这行结果怎么是空的呢?求解,多谢!
melogit reimvaluenew sex age1 age2 age3 hostype1 hostype2 hostype3 crosstype1 crosstype2 crosstype3 if crosstypenew !=4 || healthcare_connetrate:, or
note: age3 omitted because of collinearity
note: hostype3 omitted because of collinearity
note: crosstype3 omitted because of collinearity
Fitting fixed-effects model:
Iteration 0: log likelihood = -1033.1814
Iteration 1: log likelihood = -1030.167
Iteration 2: log likelihood = -1030.1549
Iteration 3: log likelihood = -1030.1549
Refining starting values:
Grid node 0: log likelihood = -1011.3995
Fitting full model:
Iteration 0: log likelihood = -1011.3995 (not concave)
Iteration 1: log likelihood = -1006.4836
Iteration 2: log likelihood = -1005.2973
Iteration 3: log likelihood = -1005.2306
Iteration 4: log likelihood = -1005.2304
Iteration 5: log likelihood = -1005.2304
Mixed-effects logistic regression Number of obs = 2259
Group variable: healthcare_c~e Number of groups = 31
Obs per group: min = 1
avg = 72.9
max = 428
Integration method: mvaghermite Integration points = 7
Wald chi2(7) = 111.23
Log likelihood = -1005.2304 Prob > chi2 = 0.0000
---------------------------------------------------------------------------------------
reimvaluenew | Odds Ratio Std. Err. z P>|z| [95% Conf. Interval]
----------------------+----------------------------------------------------------------
sex | 1.00267 .1182784 0.02 0.982 .7956968 1.263481
age1 | .0155953 .010317 -6.29 0.000 .0042647 .0570298
age2 | .0154907 .0097276 -6.64 0.000 .0045243 .0530391
age3 | 1 (omitted)
hostype1 | .8117517 .1553705 -1.09 0.276 .5578301 1.181257
hostype2 | .6259127 .1061031 -2.76 0.006 .4489731 .8725839
hostype3 | 1 (omitted)
crosstype1 | .279237 .0444772 -8.01 0.000 .2043585 .3815516
crosstype2 | .5533399 .0813931 -4.02 0.000 .4147489 .738242
crosstype3 | 1 (omitted)
_cons | 35.44314 23.66251 5.34 0.000 9.577536 131.1627
----------------------+----------------------------------------------------------------
healthcare_connetrate |
var(_cons)| .2861282 .1104249 .1342945 .6096256
---------------------------------------------------------------------------------------
LR test vs. logistic regression: chibar2(01) = 49.85 Prob>=chibar2 = 0.0000