dr_asympar.Rddr_asympar computes a normal approximation of the likelihood at a vector of threshold
values
dr_asympar(y, x, thresh, ...)
| y | outcome variable |
|---|---|
| x | matrix of covariates |
| thresh | vector of threshold values on the support of outcome y |
| ... | additional arguments to pass to |
a list of glm objects corresponding to thresh
y = faithful$waiting x = scale(cbind(faithful$eruptions,faithful$eruptions^2)) qtaus = quantile(y,c(0.05,0.25,0.5,0.75,0.95)) drabj<- dr_asympar(y=y,x=x,thresh = qtaus) lapply(drabj,coef); lapply(drabj,vcov)#> $`5%` #> (Intercept) X1 X2 #> -12.529504 7.552197 -16.412033 #> #> $`25%` #> (Intercept) X1 X2 #> -6.222253 4.537972 -10.468049 #> #> $`50%` #> (Intercept) X1 X2 #> 2.048402 -13.100002 8.855666 #> #> $`75%` #> (Intercept) X1 X2 #> 2.522160 -6.544941 4.100806 #> #> $`95%` #> (Intercept) X1 X2 #> 8.356307 -21.168439 13.933436 #>#> $`5%` #> (Intercept) X1 X2 #> (Intercept) 295.8311 -401.3021 648.1783 #> X1 -401.3021 568.4650 -904.5508 #> X2 648.1783 -904.5508 1446.7640 #> #> $`25%` #> (Intercept) X1 X2 #> (Intercept) 8.957952 -16.91239 24.55477 #> X1 -16.912395 40.88275 -55.52501 #> X2 24.554773 -55.52501 76.72885 #> #> $`50%` #> (Intercept) X1 X2 #> (Intercept) 0.9382644 -5.362741 4.168658 #> X1 -5.3627412 33.586619 -26.632828 #> X2 4.1686583 -26.632828 21.260327 #> #> $`75%` #> (Intercept) X1 X2 #> (Intercept) 0.3778717 -1.93365 1.464948 #> X1 -1.9336498 13.05132 -10.482798 #> X2 1.4649478 -10.48280 8.539182 #> #> $`95%` #> (Intercept) X1 X2 #> (Intercept) 36.33379 -165.7202 120.2247 #> X1 -165.72018 773.3070 -564.5566 #> X2 120.22469 -564.5566 412.9514 #># mean and covariance at respective threshold values