The supposition seems to be that
1. Some people are reluctant to do the work needed to find out for
themselves because that entails using two pieces of software, namely
Stata and a browser to search the Statalist archives.
2. So, let Stata add the functionality of searching the Statalist
archives. Then you can do it all in Stata.
I am not against adding functionality to Stata but I don't buy this one.
I think we have a fairly mundane human problem, i.e. some people are too
lazy or (more politely) too busy to follow all the advice given to them.
Often people would search if only they knew the right keywords, but they
don't, which is why they are asking people. Otherwise
I think there are fairly mundane human solutions to this, namely
(a) promoting a friendly but not fickle atmosphere on the list,
including (as Eva Poen memorably put it a recent thread) saying
occasionally "Come on, you can figure that one out by yourself".
(b) reaching for Delete whenever it is what people seem to deserve. So,
for example, some people go on fishing expeditions to see what they can
they pick up from Statalist, and that's not outrageous, but many of us
can quite reasonably decide that it's a waste of our time to give them
much support.
Nick
[email protected]
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Martin Weiss
Sent: 02 July 2008 14:50
To: [email protected]
Subject: RE: R: st: RE: montecarlo simulations with actual distributions
Regarding unnecessary queries to the list: how much would it take to
implement a search function for the Statalist discussions in official
Stata`s help menu? -search- could automatically include the material
available from the "Archive search" button on the site
http://www.stata.com/statalist/archive/. That would surely lower the
frequency of unnecessary posts...
Martin Weiss
_________________________________________________________________
Diplom-Kaufmann Martin Weiss
Mohlstrasse 36
Room 415
72074 Tuebingen
Germany
Fon: 0049-7071-2978184
Home: http://www.wiwi.uni-tuebingen.de/cms/index.php?id=1130
Publications: http://www.wiwi.uni-tuebingen.de/cms/index.php?id=1131
SSRN: http://papers.ssrn.com/sol3/cf_dev/AbsByAuth.cfm?per_id=669945
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Nick Cox
Sent: Wednesday, July 02, 2008 3:25 PM
To: [email protected]
Subject: RE: R: st: RE: montecarlo simulations with actual distributions
Please see advice in the Statalist FAQ on how to find out things. The
list is not a substitute for -help-, -search-, -findit- or Google.
You can do nonparametric regressions in various ways in Stata.
Official commands include -lowess-, -lpoly-.
Unofficial commands include -mrunning-, -rcspline-.
Outdated unofficial commands include anything whose name starts with
-kernreg-.
Use -findit- for precise locations.
However, nonparametric regression with multiple predictors is not well
supported in Stata -- except that, arguably, fractional polynomials are
very well supported.
More to the point, I am not clear how you would combine bootstrapping
with any regression not parametrically specified. What do you have in
mind?
Nick
[email protected]
emanuele canegrati
I am performing the bootstrap regressions. Do you know which is the
STATA command to perform non-parametric bootstrap regressions?
*
* 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/
*
* 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/
*
* 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/
