---------------------------------------------------------------------------------
       log:  C:\Documents and Settings\agailey\Desktop\hyp_mata\wealth_subsamp_ah
> g.log
  log type:  text
 opened on:  17 Jun 2008, 17:00:30

. 
. set more off

. set mem 100m
(102400k)

. 
. use hatota black hispan male hsless college widowed using \\homer\homer_c\Retir
> e\ahg\age5055_hrs1992r_v9.dta , clear

. drop if hatota > 500 | hatota < -100
(289 observations deleted)

. ghreg_ahg2 hatota black hispan male hsless college widowed

Maximizing concentrated likelihood: 
Iteration 0:  f(p) = -29797.344
Iteration 1:  f(p) = -29702.683
Iteration 2:  f(p) = -29639.063
Iteration 3:  f(p) = -29329.171
Iteration 4:  f(p) = -29204.547
Iteration 5:  f(p) = -29087.869
Iteration 6:  f(p) = -29087.869
Iteration 7:  f(p) = -29046.579
Iteration 8:  f(p) = -29046.579
Iteration 9:  f(p) = -29046.579
Iteration 10:  f(p) = -29046.579
Iteration 11:  f(p) = -29046.579
Iteration 12:  f(p) = -29046.407
Iteration 13:  f(p) = -29046.407
Iteration 14:  f(p) = -29046.407
Iteration 15:  f(p) = -29046.407
Iteration 16:  f(p) = -29046.406
Iteration 17:  f(p) = -29046.406

Retrieving concentrated parameters: 

. 
. 
. predict yhat3, simu

. 
. 
. **Compare simulated and actual.**
. summ hatota, d

                      HH wlth in 1000s
-------------------------------------------------------------
      Percentiles      Smallest
 1%          -24          -99.7
 5%          -.3            -98
10%         .675      -79.48761       Obs                3447
25%           21          -75.5       Sum of Wgt.        3447

50%       71.548                      Mean           104.2378
                        Largest       Std. Dev.      108.0413
75%        155.2            495
90%     264.8571          496.3       Variance       11672.92
95%          340       497.1382       Skewness       1.294198
99%     442.9764       497.1382       Kurtosis       4.256691

. drop if yhat3 >= r(p99)
(136 observations deleted)

. 
. sum yhat3 hatota, detail

                    Simulated Value of Y
-------------------------------------------------------------
      Percentiles      Smallest
 1%    -34.45448      -246.4693
 5%     -7.45883      -200.5511
10%     3.668427      -101.7826       Obs                3311
25%     25.86062      -92.94951       Sum of Wgt.        3311

50%     60.39904                      Mean           87.78217
                        Largest       Std. Dev.      91.19792
75%      123.653       440.5745
90%     215.0062       441.0702       Variance        8317.06
95%     281.6618       442.6719       Skewness       1.429189
99%     403.1109       442.9413       Kurtosis       5.128375

                      HH wlth in 1000s
-------------------------------------------------------------
      Percentiles      Smallest
 1%        -20.4          -99.7
 5%          -.3            -98
10%           .5      -79.48761       Obs                3311
25%         20.2          -75.5       Sum of Wgt.        3311

50%         70.7                      Mean           102.5483
                        Largest       Std. Dev.      106.7915
75%        153.8          492.2
90%          258          496.3       Variance       11404.43
95%          339       497.1382       Skewness       1.310915
99%        436.5       497.1382       Kurtosis       4.326595

. 
. twoway (kdensity yhat3 if yhat3<500) (kdensity hatota if hatota<500), ti("Compa
> rison of Simulated and Actual Wealth")

. graph export simu_subsamp_wealth.png, as(png) replace
(file simu_subsamp_wealth.png written in PNG format)

. 
. 
. keep hatota yhat3

. 
. 
. 
. gen id = _n

. 
. rename hatota w1

. rename yhat3 w2

. reshape long w, i(id) j(source)
(note: j = 1 2)

Data                               wide   ->   long
-----------------------------------------------------------------------------
Number of obs.                     3311   ->    6622
Number of variables                   3   ->       3
j variable (2 values)                     ->   source
xij variables:
                                  w1 w2   ->   w
-----------------------------------------------------------------------------

. 
. ksmirnov w, by(source)

Two-sample Kolmogorov-Smirnov test for equality of distribution functions

 Smaller group       D       P-value  Corrected
 ----------------------------------------------
 1:                  0.0586    0.000
 2:                 -0.0731    0.000
 Combined K-S:       0.0731    0.000      0.000

Note: ties exist in combined dataset;
      there are 4816 unique values out of 6622 observations.

. 
. log close
       log:  C:\Documents and Settings\agailey\Desktop\hyp_mata\wealth_subsamp_ah
> g.log
  log type:  text
 closed on:  17 Jun 2008, 17:01:28
---------------------------------------------------------------------------------