Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: bootstrap statistic

From   "Nick Winter" <>
To   <>
Subject   st: RE: bootstrap statistic
Date   Thu, 12 Sep 2002 10:22:22 -0400

> -----Original Message-----
> From: Jun Xu [] 
> Sent: Thursday, September 12, 2002 10:15 AM
> To:
> Subject: st: bootstrap statistic
> In bs command, Stata puts a formula in the second set of 
> quotation marks to 
> denote a statistic to bootstrap like
> logit y x1 x2
> bs "logit y x1 x2" "whatever statistic formula", rep(#)
> However, it's kind of strange to me that I cannot wrap up 
> lines use /* */ 
> and it seems that I cannot put space between numeric signs 
> like "_b[_cons] + 
> mean2*_b[x1] + mean3*_b[_x2]" won't work, and I have to get 
> rid of any space 
> like "_b[_cons]+mean2*_b[x1]+mean3*_b[_x2]".  Here mean2 and 
> mean3 are 
> defined as scalars.  Any input on this?  Is that because I 
> use scalar in the 
> quotation mark, or it has to be the case that no wrap-up of 
> the stuff within 
> the second set of quotation marks or the space thing.  If so, 
> might be kind 
> of inconvient if I have a long formula for a statistic to bootstrap.

The -bs- command breaks up the line of statistics with the -tokenize-
command, splitting the line on white space.  You can include spaces in
the individual statistics by enclosing them in quotation marks, and
enclosing the whole list in compound quotes:

bs "reg mpg price weight" `""_b[price] + _b[weight]" "_b[price]"

However, as far as I can tell you cannot break up the line of statistics
two lines.

Nick Winter

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index