How to capture information generated by the command "hoi"

[Girlan Oliveira]

Hi Statalisters,

I am using the "hoi" command to calculate the Human Opportunity Index for 27 states and 2 years and to get the decomposition of the Index for each state comparing the two years. The code I'm using is as follows:

local predictors presmae metrop area logrenpcdef nmorad refsexo refraca medescresp difescresp

local outcomes agua

levelsof uf, local(ufs)
foreach c of local ufs {
display "uf = `c'"
foreach o of varlist `outcomes' {
capture noisily hoi `o' `predictors' [fw = pesopes] ///
if uf == `c', by (ano) format(%9.3f) estimates decomp2
if c(rc) == 2000 { // hoi FAILED DUE TO NO OBSERVATIONS
display "Nao ha observacoes ou o outcome nao e dicotomico `o': analise ignorada"
}
else if c(rc) != 0 { // SOME OTHER ERROR AROSE ATTEMPTING hoi
display in red "Erro encontrado ao executar o ioh com o outcome `o', uf = `c', ano = `y'"
exit c(rc) // SHOW ERROR CODE AND STOP
}

}
}

The output of Stata to a state is as follows:

uf = 53

(sum of wgt is   3,7435e+05)
note: metrop dropped because of collinearity
Iteration 0:   log pseudolikelihood = -555,11988
Iteration 1:   log pseudolikelihood = -549,95284
Iteration 2:   log pseudolikelihood = -549,63683
Iteration 3:   log pseudolikelihood = -549,63609

Logistic regression                               Number of obs   =       1724
                                                  Wald chi2(8)    =      15,00
                                                  Prob > chi2     =     0,0590
Log pseudolikelihood = -549,63609                 Pseudo R2       =     0,0099

------------------------------------------------------------------------------
             |               Robust
        agua |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
     presmae |  -,2627851    ,473094    -0,56   0,579    -1,190032    ,6644622
        area |   ,7122128    ,242673     2,93   0,003     ,2365825    1,187843
 logrenpcdef |  -,0157898   ,0214751    -0,74   0,462    -,0578802    ,0263006
      nmorad |   ,0402422   ,0552911     0,73   0,467    -,0681263    ,1486107
     refsexo |   ,3394054   1,127855     0,30   0,763    -1,871149     2,54996
     refraca |   ,1862689   ,1616822     1,15   0,249    -,1306224    ,5031603
  medescresp |   -,018722   ,0228352    -0,82   0,412    -,0634781    ,0260342
  difescresp |   ,0011085   ,0337969     0,03   0,974    -,0651322    ,0673493
       _cons |   1,425572   1,415576     1,01   0,314    -1,348907    4,200051
------------------------------------------------------------------------------

(sum of wgt is   3,7435e+05)
note: metrop dropped because of collinearity
Iteration 0:   log pseudolikelihood = -555,11988
Iteration 1:   log pseudolikelihood = -549,95284
Iteration 2:   log pseudolikelihood = -549,63683
Iteration 3:   log pseudolikelihood = -549,63609

Logistic regression                               Number of obs   =       1724
                                                  Wald chi2(8)    =      15,00
                                                  Prob > chi2     =     0,0590
Log pseudolikelihood = -549,63609                 Pseudo R2       =     0,0099

------------------------------------------------------------------------------
             |               Robust
        agua |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
     presmae |  -,2627851    ,473094    -0,56   0,579    -1,190032    ,6644622
        area |   ,7122128    ,242673     2,93   0,003     ,2365825    1,187843
 logrenpcdef |  -,0157898   ,0214751    -0,74   0,462    -,0578802    ,0263006
      nmorad |   ,0402422   ,0552911     0,73   0,467    -,0681263    ,1486107
     refsexo |   ,3394054   1,127855     0,30   0,763    -1,871149     2,54996
     refraca |   ,1862689   ,1616822     1,15   0,249    -,1306224    ,5031603
  medescresp |   -,018722   ,0228352    -0,82   0,412    -,0634781    ,0260342
  difescresp |   ,0011085   ,0337969     0,03   0,974    -,0651322    ,0673493
       _cons |   1,425572   1,415576     1,01   0,314    -1,348907    4,200051
------------------------------------------------------------------------------
WARNING :  ano = 95 : 1 initially selected circunstances were not used to estimate Pi.

note: refsexo != 1 predicts success perfectly
      refsexo dropped and 101 obs not used

(sum of wgt is   3,7210e+05)
note: metrop dropped because of collinearity
Iteration 0:   log pseudolikelihood = -127,39502
Iteration 1:   log pseudolikelihood = -106,99904
Iteration 2:   log pseudolikelihood = -100,02991
Iteration 3:   log pseudolikelihood = -97,741978
Iteration 4:   log pseudolikelihood = -97,334723
Iteration 5:   log pseudolikelihood = -97,325352
Iteration 6:   log pseudolikelihood = -97,325344

Logistic regression                               Number of obs   =       1798
                                                  Wald chi2(7)    =      28,18
                                                  Prob > chi2     =     0,0002
Log pseudolikelihood = -97,325344                 Pseudo R2       =     0,2360

------------------------------------------------------------------------------
             |               Robust
        agua |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
     presmae |   ,4367261   ,9265243     0,47   0,637    -1,379228     2,25268
        area |   ,2214284   ,6466493     0,34   0,732    -1,045981    1,488838
 logrenpcdef |   ,3554017   ,2264125     1,57   0,116    -,0883587    ,7991622
      nmorad |  -,3045169   ,1124967    -2,71   0,007    -,5250063   -,0840275
     refraca |  -,9231166    ,485408    -1,90   0,057    -1,874499    ,0282656
  medescresp |   ,3880012   ,1011046     3,84   0,000     ,1898399    ,5861625
  difescresp |  -,1500026   ,0815967    -1,84   0,066    -,3099292    ,0099239
       _cons |   2,616778   1,246677     2,10   0,036     ,1733361    5,060221
------------------------------------------------------------------------------

(sum of wgt is   3,7435e+05)
note: metrop dropped because of collinearity
Iteration 0:   log pseudolikelihood = -555,11988
Iteration 1:   log pseudolikelihood = -549,95284
Iteration 2:   log pseudolikelihood = -549,63683
Iteration 3:   log pseudolikelihood = -549,63609

Logistic regression                               Number of obs   =       1724
                                                  Wald chi2(8)    =      15,00
                                                  Prob > chi2     =     0,0590
Log pseudolikelihood = -549,63609                 Pseudo R2       =     0,0099

------------------------------------------------------------------------------
             |               Robust
        agua |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
     presmae |  -,2627851    ,473094    -0,56   0,579    -1,190032    ,6644622
        area |   ,7122128    ,242673     2,93   0,003     ,2365825    1,187843
 logrenpcdef |  -,0157898   ,0214751    -0,74   0,462    -,0578802    ,0263006
      nmorad |   ,0402422   ,0552911     0,73   0,467    -,0681263    ,1486107
     refsexo |   ,3394054   1,127855     0,30   0,763    -1,871149     2,54996
     refraca |   ,1862689   ,1616822     1,15   0,249    -,1306224    ,5031603
  medescresp |   -,018722   ,0228352    -0,82   0,412    -,0634781    ,0260342
  difescresp |   ,0011085   ,0337969     0,03   0,974    -,0651322    ,0673493
       _cons |   1,425572   1,415576     1,01   0,314    -1,348907    4,200051
------------------------------------------------------------------------------
WARNING :  ano = 2002 : 2 initially selected circunstances were not used to estimate Pi.

--------------------------------------------------------------------------
                              |                    Type                   
         By: ano and Variable |    Values  Std Error    LB (95)    UB (95)
------------------------------+-------------------------------------------
95                            |
                 Coverage (C) |    90,140      0,716     88,737     91,543
            Dissemilarity (D) |     0,907      1,242     -1,527      3,341
Human Opportunity Index (HOI) |    89,322      0,809     87,736     90,908
                    Pseudo R2 |     0,010                                 
                    Obs Logit |  1724,000                                 
                          Obs |  1724,000                                 
                      Wtg Pop |  3,74e+05                                 
               Vulnerable Pop |  1,29e+05                                 
               Vulnerable (%) |    34,514                                 
                        Obs 1 |  1724,000                                 
                        Obs 2 |  1724,000                                 
                     Loss (%) |     0,000                                 
------------------------------+-------------------------------------------
2002                          |
                 Coverage (C) |    98,666      0,253     98,170     99,161
            Dissemilarity (D) |     0,811      0,916     -0,985      2,608
Human Opportunity Index (HOI) |    97,865      0,405     97,071     98,659
                    Pseudo R2 |     0,236                                 
                    Obs Logit |  1798,000                                 
                          Obs |  1798,000                                 
                      Wtg Pop |  3,72e+05                                 
               Vulnerable Pop | 93747,000                                 
               Vulnerable (%) |    25,194                                 
                        Obs 1 |  1899,000                                 
                        Obs 2 |  1899,000                                 
                     Loss (%) |     0,000                                 
--------------------------------------------------------------------------

------------------------------------------------
Decomposition 2: Composition, |
Equalizatio and Scale effects |     By: ano     
and Variable                  |      95     2002
------------------------------+-----------------
Original                      |
                 Coverage (C) |  90,140   98,666
            Dissemilarity (D) |   0,907    0,811
Human Opportunity Index (HOI) |  89,322   97,865
------------------------------+-----------------
Decomposition (p.p.)          |
                Change (p.p.) |            8,543
           Composition (p.p.) |           -0,371
                 Scale (p.p.) |            8,840
          Equalization (p.p.) |            0,074
------------------------------+-----------------
Decomposition (%)             |
              Composition (%) |           -4,341
                    Scale (%) |          103,478
             Equalization (%) |            0,863
------------------------------------------------

I would like to know how to modify my code in order to capture the following informations:

1 - Coverage (C)
2 - Dissemilarity (D)
3 - Human Opportunity Index (HOI)
4 - Change (p.p.)
5 - Composition (p.p.)
6 - Scale (p.p.)
7 - Equalization (p.p.)
5 - Composition (%)
6 - Scale (%)
7 - Equalization (%)


Thanks in advance.

[Nick Cox]

Quoting from FAQ Advice #12

If you are using user-written commands, explain that and say where they came from: the Stata Journal, SSC, or other archives. This helps (often crucially) in explaining your precise problem, and it alerts readers to commands that may be interesting or useful to them.

Here are some examples:

I am using xtreg in Stata 13.1.

I am using estout from SSC in Stata 13.1.

I did a search hoi and got 80 hits, many on choice, etc. At that point I gave up.

It really is your job to tell us where this program comes from.

[Girlan Oliveira]

Nick,

I'm using hoi from SSC in Stata 12

[Nick Cox]

That's helpful. All I can see in the help is

Saved Results

hoi return results in r() format. Type return list after estimation.

That's not very helpful. The clear standard here is a list of named results with statements of what they are.

At this point, I bail out as I don't have the inclination to read the help, read the code, or play with the command, which is far from trivial. So I can't even guess whether (a) what you want is easy because your shopping list is matched by saved results, or (b) your shopping list implies a substantial programming job.

Others may be able to help (much) more.

[Bruce Weaver]

I would try asking for support via the e-mail address shown in the output from:

ssc describe hoishapely

HTH.

[Tista Kundu]

I am using hoishapley decomposition but it shows this massage
help hoishapley

. hoishapley hhop [fw=weights] if year==2006, shapley( psch light santi water)
weights not found
r(111);
when i remove weight

. hoishapley hhop if year==2006, shapley( psch light santi water)
unrecognized command: catenate
r(199);

- One needs to install package catenate

ssc install catenate

[Shahla Akram]

anyone can help me to run inter generational mobility and intra generational mobility in different way as relative, absolute, decomposition, upper and lower mobility in stata 13.
thank you