Paul,
1. You probably need to increase quantile.
2. There is brief discussion of bias in [R] bootstrap, page 150.
You also might the following program useful; unlike -clad- or -cqreg- it
only bootstraps (as an option) the standard errors, not the model
coefficients.
Scott
program cqreg2
version 9.2
syntax varlist(min=1) [if] [, Quantile(real 0.5) ll(string)
ul(string) bs reps(integer 100) save(string) grqreg]
tempvar yhat
tokenize `varlist'
local lhs "`1'"
mac shift
local rhs "`*'"
preserve
qui {
qreg `lhs' `rhs' `if', q(`quantile')
predict `yhat'
sum `hat'
tempvar tag
if "`ll'" != "" & "`ul'" == "" {
while r(min) < `ll' {
keep if `yhat' >= `ll'
qui qreg `lhs' `rhs' `if', quantile(`quantile' )
qui drop `yhat'
predict `yhat' if e(sample)
qui sum `yhat' if e(sample)
}
}
if "`ll'" == "" & "`ul'" != "" {
while r(max) > `ul' {
keep if `yhat' <= `ul'
qui qreg `lhs' `rhs' `if', quantile(`quantile' )
qui drop `yhat'
predict `yhat' if e(sample)
qui sum `yhat' if e(sample)
}
}
if "`ll'" != "" & "`ul'" != "" {
while r(max) > `ul' & r(min) < `ll' {
keep if `yhat' <= `ul' & `yhat' >= `ll'
qui qreg `lhs' `rhs' `if', quantile(`quantile' )
qui drop `yhat'
predict `yhat' if e(sample)
qui sum `yhat' if e(sample)
}
}
}
if "`bs'" == "bs" {
bsqreg `lhs' `rhs' `if', quantile(`quantile' ) reps(`reps')
}
else {
qreg
}
if "`grqreg'" != "" {
grqreg, ci
}
if "`save'" != "" {
disp ""
save "`save'", replace
}
end
> -----Original Message-----
> From: [email protected] [mailto:owner-
> [email protected]] On Behalf Of Paul GERRANS
> Sent: Thursday, September 07, 2006 8:35 PM
> To: [email protected]
> Subject: st: CQREG CLAD
>
> Hello,
> I am using Professor Ken Chay's censored regression programs (cqreg and
> scls) available from his website http://elsa.berkeley.edu/~kenchay/
> Two questions:
> - While I can run Tobit and SCLS and get estimates most samples where I
> run cqreg i get "No covergence" after some considerable time. I have
> approx 4800 left-censored observations (0), 11200 uncensored observations
> and 1870 right-censored (100)
> - When I do get output for the cqreg and scls estimators what is the
> interpretation of the Bias output? Do I add/subtract from the coefficient
> estimate?
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
