Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: stratification matching, please advice!!


From   "Jenniffer Solorzano Mosquera" <[email protected]>
To   <[email protected]>
Subject   st: stratification matching, please advice!!
Date   Sun, 20 May 2012 13:20:59 +0200

Hi everybody, please help!!!!


I have a question that I would appreciate you can give some hints.I need to run stratification matching and then present pre-treatment outcomes ttests after matching by quintiles of pscore, to check if the pre_treatment outcomes were actually balanced. What I did was :	

1. -pscore PE  _IUO_1 _Ilisted_1 _IUOXlis_1_1  ent1_bGearing   ln_ent1_bTotalassetsthEUR change_cash_flow Age Age2 ent1_bCF_Ope _Iyear_entr_*, pscore(mypscore2) blockid(myquintiles2) comsup level(0.005) -


2.
After obtaining the pscore quantiles (myquintiles2 actually gave me 5 groups) I run the following for stratification matching:

  -      g att = .
  g seatt=.
  g weight=.
  g support=.
  g treated=.
        egen g1 = group(myquintiles2) // listed , NACE_1digit_3 , year_entry pscore_q2
        levels g1, local(gr)
        qui foreach j of local gr {
    psmatch2 PE  ///
    if g1==`j', pscore(pscore_2) out(exi1_aEBITDA_multiple) common
                replace att = r(att) if  g1==`j' & e(sample) 
    replace seatt = r(seatt) if g1==`j' & e(sample) 
    replace weight=_weight if g1==`j' & e(sample) 
    replace support=_support if g1==`j' & e(sample) 
    replace treated=_treated if g1==`j' & e(sample) -
   


3. And then I try to test if after matching the difference in means of pre-acquisition  outcome is not significant between treated and non-treated:

-  levelsof pscore_2_q, local(gg)
  foreach i of local gg {
preserve
keep if g1==`i' & support==1
pstest  ent1_bEBITDA_multiple , treated(treated) support(support) mweight(weight)
restore
} -

Is it ok? How do I test whether the ATT was significant or not considering that it was with stratification matching?.

Thanks a lot for your advice..





*
*   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index