Re: st: Manually compute IPTW for PS matching?

[TA Stat <tastat@gmail.com>]

Thanks for your reply.  Actually, I did compute iptw manually already.
 I am not sure whether I do need iptw with 1:1 matching, but not 1:n
matching.  I am comparing 3 levels variable, (treatment A vs C and B
vs C) on survival events by Cox PH model.  I am not quite sure about
the weight that Stata produce with 1:1 matching.  That is the reason
why I get confused.

Thanks
Pete

On Fri, Mar 16, 2012 at 9:51 PM, Adam Olszewski
<adam.olszewski@gmail.com> wrote:
> Hi Pete,
> Are you doing PS matching or weighting? If you're only doing 1:1
> matching then IPTW is not necessary.
> I usually calculate IPTW manually:
>
> gen iptw = treatment/pscore + (1-treatment)/(1-pscore)
>
> I am not aware of any "step-by-step" Stata-specific tutorial on PS
> matching. It is a complex process if to be done properly, with
> appropriate diagnostics and many potential pitfalls. One brief general
> intro is at:
>
> http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3144483/
>
> Also, read the help for -pscore2- -pstest- -psgraph- and -pbalchk- thoroughly.
>
> Adam Olszewski
>
> On Fri, Mar 16, 2012 at 2:03 AM, TA Stat <tastat@gmail.com> wrote:
>> Dear All
>>
>> This is a simple question.  After run psmatch2 command (1:1 matching),
>> I got pscore, treated, support, weight, n1, nn, pdif variables.  Do I
>> have to manually calculate IPTW? Is any recommended step-by-step
>> illustration of PS matching in Stata available?
>>
>> Thanks
>> Pete
>> *
>> *   For searches and help try:
>> *   http://www.stata.com/help.cgi?search
>> *   http://www.stata.com/support/statalist/faq
>> *   http://www.ats.ucla.edu/stat/stata/
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/