The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.
Regression models for limited dependent variables are interpreted in several different ways. The estimated coefficients are often transformed before being interpreted, for example. Exponentiating logistic or count regression coefficients is common, as is converting the coefficient into percent change. The list.coef function transforms model coefficients into these additional quantities. It also reports confidence intervals for exponentiated coefficients. Users may specify the # of decimal places to report and the alpha level. For example:
library(catregs)
data("Mize19AH")
m1 <- glm(alcB ~woman*parrole + age + race2 + race3 + race4 + income + ed1 + ed2 + ed3 + ed4,family="binomial",data=Mize19AH)
list.coef(m1)## variables b SE z ll ul p.val exp.b ll.exp.b
## 1 (Intercept) 2.627 0.544 4.827 1.560 3.694 0.000 13.836 4.761
## 2 woman -0.284 0.095 -2.999 -0.470 -0.098 0.004 0.753 0.625
## 3 parrole -0.482 0.105 -4.568 -0.688 -0.275 0.000 0.618 0.502
## 4 age -0.086 0.019 -4.603 -0.123 -0.049 0.000 0.917 0.884
## 5 race2 -0.445 0.078 -5.726 -0.598 -0.293 0.000 0.641 0.550
## 6 race3 0.284 0.443 0.641 -0.584 1.152 0.325 1.328 0.557
## 7 race4 -0.153 0.197 -0.778 -0.539 0.233 0.295 0.858 0.583
## 8 income 0.008 0.001 5.505 0.005 0.011 0.000 1.008 1.005
## 9 ed1 0.460 0.153 3.006 0.160 0.760 0.004 1.584 1.173
## 10 ed2 0.610 0.143 4.256 0.329 0.891 0.000 1.840 1.390
## 11 ed3 0.765 0.154 4.967 0.463 1.067 0.000 2.150 1.589
## 12 ed4 0.680 0.177 3.833 0.332 1.028 0.000 1.974 1.394
## 13 woman:parrole -0.215 0.138 -1.561 -0.486 0.055 0.118 0.806 0.615
## ul.exp.b percent CI
## 1 40.205 1283.553 95 %
## 2 0.906 -24.723 95 %
## 3 0.760 -38.228 95 %
## 4 0.952 -8.253 95 %
## 5 0.746 -35.933 95 %
## 6 3.166 32.842 95 %
## 7 1.262 -14.202 95 %
## 8 1.011 0.798 95 %
## 9 2.137 58.371 95 %
## 10 2.437 84.042 95 %
## 11 2.908 114.969 95 %
## 12 2.795 97.413 95 %
## 13 1.057 -19.368 95 %In the above, b refers to the original model coefficient, S.E. to the estimated model coefficient’s standard error, Z is the z-statistic (b/se(b)), LL.CI is the lower limit of the confidence interval around the model coefficient, UL.CI is the upper limit of the confidence interval around the model coefficient, p-val is the p-value associated with the z-statistic, exp.b is the exponentiated model coefficient, LL.CI.for.exp.b is the lower limit of the confidence interval around the exponentiated model coefficient, UL.CI.for.exp.b is the upper limit of the confidence interval around the exponentiated model coefficient, Percent is the % change interpretation of the coefficient, i.e., 100*(1-(exp(b))), and CI is the amount of confidence in the confidence interval.
These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.