* ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~--;


 /*
 *  13 June 2006
 
CEO in Japan project

Table 6

run by Eric. 

In converting exec7 to stata 9, I lost all the uppercase in variable names, so I put everything below in lowercase. 

Before running this, make sure that exec7.dta and temp.txt (which can be a blank file) exist in the same directory as this DO file. 

 */

 * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~--;
  #delimit ;
 * This says that the semicolon denotes the end of a line of command.
All lines must end with semicolons after this;

*log using temp.log, replace;

set more 1;
 *This should stop the pauses;
 *To keep going regardless of errors, use DO mYFILE, nostop;
   * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~--;


set memory 50m;

set matsize 120; 

 * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~--;
 
 

use  exec7.dta;

 
 

* ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~--;
  

 
tobit  lntax04tb  lnassets opin04cap grsp0403jun secactg  indus1-indus31  fatcat lnassfc  profitfc spjunefc if prez == 1, ll;
 
mfx compute, predict (ys(9.2103405,.)) at (median);

 outreg using temp.txt,  replace  bdec(2)   coefastr  3aster comma; 


  * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~--;

  tobit  lntax04tb  lnassets opin04cap grsp0403jun multipos espall optnprgm  offage05 secactg  indus1-indus31  fatcat lnassfc  profitfc spjunefc   mulposfc 
  optnpmfatcat  excagefc if prez == 1, ll;

 

  mfx compute, predict (ys(9.2103405,.)) at (median);

outreg using temp.txt, append  bdec(2)   coefastr  3aster comma; 

  * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~--;

     /*  
 	Now for the question of what to include for Column (3).  I ran a zillion iterations trying to figure out what variables were messing up the convergence.  It seemed to me that the problem law in our Committee and interacted Committee variable.  I couldn't get the regressions with an interacted Committee variable to converge.  
	What I'd suggest, then, is that somewhere we note this problem -- and then simply discuss the effect of the Committee variable on the whole sample.  For Table 3, we ignore the committee variable and present the following regression instead:
    */
 

  tobit  lntax04tb  lnassets opin04cap grsp0403jun multipos espall optnprgm  offage05 ownfamco othfamco  othbdsh ftshcorp sh1  tot5sh tot10sh bdsize inddrpt
   bdtenure bdage secactg  indus1-indus31  fatcat lnassfc  profitfc spjunefc   
  mulposfc optnpmfatcat  excagefc  othfamfc othbdshfc ftshcofc topshfc top5fc top10fc bdszfc indrptfc bdtenfc bdagefc if prez == 1, ll;

 
  mfx compute, predict (ys(9.2103405,.)) at (median);

outreg using temp.txt, append  bdec(2)   coefastr  3aster comma; 
 

 * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~--;

*log close;

clear;

 * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~--;