r - how to report goodness of fit for a svyglm object? -
a question survey package in r. know basic i've found nothing. so:
library(survey) data(api) dclus2 <- svydesign(id=~dnum, weights=~pw, data=apiclus1, fpc=~fpc) summary(svyglm(api00 ~ ell + meals + mobility, design = dclus2)) call: svyglm(formula = api00 ~ ell + meals + mobility, design = dclus2) survey design: svydesign(id = ~dnum, weights = ~pw, data = apiclus1, fpc = ~fpc) coefficients: estimate std. error t value pr(>|t|) (intercept) 819.2791 21.3900 38.302 4.65e-13 *** ell -0.5167 0.3240 -1.595 0.139 meals -3.1232 0.2781 -11.231 2.29e-07 *** mobility -0.1689 0.4449 -0.380 0.711 --- signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 (dispersion parameter gaussian family taken 3157.85) number of fisher scoring iterations: 2
is there r-squared or other statistic give reader impression of how model fits data?
thanks.
there package came out in late 2016 found called "poliscidata" has function fit.svyglm returns r square , adj. r square outputs svyglm objects. have used success.
Comments
Post a Comment