-----------------------------------------------------------------------------------------------------

name:  <unnamed>

log:  C:\Users\mexmi\Documents\newer web pages\Soc_382\logs\2nd loglin intermar class.log

log type:  text

opened on:  31 Jan 2019, 10:18:08

* Before class I was creating the crossings parameters, which I did not get to explain in class (and will next class).

. * replace cross1=1 if (meth_num==1 & feth_num>1) | (feth_num==1& meth_num>1)

. gen cross1=0

. replace cross1=1 if (meth_num==1 & feth_num>1) | (feth_num==1& meth_num>1)

. gen cross2=0

. replace cross2=1 if (meth_num<=2 & feth_num>2)| (feth_num<=2 & meth_num>2)

. gen cross3=0

. replace cross3=1 if (meth_num<=3 & feth_num>3)| (feth_num<=3 & meth_num>3)

. gen cross4=0

. replace cross4=1 if (meth_num<=4 & feth_num>4)| (feth_num<=4 & meth_num>4)

. poisson count i.meth_num i.feth_num i.cross*

Iteration 0:   log likelihood = -1628492.6

Iteration 1:   log likelihood = -514542.18  (backed up)

Iteration 2:   log likelihood = -405521.33  (backed up)

Iteration 3:   log likelihood = -241959.01  (backed up)

Iteration 4:   log likelihood = -147645.83

Iteration 5:   log likelihood = -59259.102

Iteration 6:   log likelihood = -12308.206

Iteration 7:   log likelihood = -10257.604

Iteration 8:   log likelihood =  -10246.31

Iteration 9:   log likelihood = -10246.309

Poisson regression                                Number of obs   =         25

LR chi2(12)     = 3132470.69

Prob > chi2     =     0.0000

Log likelihood = -10246.309                       Pseudo R2       =     0.9935

-------------------------------------------------------------------------------------

count |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]

--------------------+----------------------------------------------------------------

meth_num |

Mexican American  |   -.977803   .0181359   -53.92   0.000    -1.013349   -.9422572

Hispanic Other  |  -1.938943    .020284   -95.59   0.000    -1.978698   -1.899187

Non Hispanic Other  |  -2.506451   .0216409  -115.82   0.000    -2.548866   -2.464035

White non Hispanic  |   .6765227   .0185343    36.50   0.000     .6401961    .7128493

|

feth_num |

Mexican American  |  -.0199891    .018195    -1.10   0.272    -.0556507    .0156725

Hispanic Other  |  -.8181779   .0202753   -40.35   0.000    -.8579168    -.778439

Non Hispanic Other  |  -1.107257   .0209967   -52.73   0.000     -1.14841   -1.066104

White non Hispanic  |   1.871097   .0185343   100.95   0.000      1.83477    1.907423

|

1.cross1 |  -3.026666    .017803  -170.01   0.000    -3.061559   -2.991773

1.cross2 |  -1.014698   .0101682   -99.79   0.000    -1.034627   -.9947684

1.cross3 |  -.5411649   .0113079   -47.86   0.000     -.563328   -.5190018

1.cross4 |  -1.444942   .0092927  -155.49   0.000    -1.463155   -1.426728

_cons |   10.65775   .0048495  2197.69   0.000     10.64825    10.66726

-------------------------------------------------------------------------------------

. poisgof

Deviance goodness-of-fit =  20262.62

Prob > chi2(12)          =    0.0000

Pearson goodness-of-fit  =  21598.27

Prob > chi2(12)          =    0.0000

* We started class here, talking about where the Quasi-independence model fit the data, and where it didn’t.

. table meth_num feth_num, contents(sum count sum quasi_indep_model) row col cellwidth(10)

--------------------------------------------------------------------------------------------

husband's           |                         wife's race/ethnicity

race/ethnicity      | Black, non  Mexican Am  Hispanic O  Non Hispan  White non        Total

--------------------+-----------------------------------------------------------------------

Black, non Hispanic |      42521         291         412         393        2064       45681

|      42521    274.1622    229.8974    264.6786    2391.262       45681

|

Mexican American |         94       18088         612         433        6067       25294

|   108.2118       18088    565.4383    650.9836    5881.366       25294

|

Hispanic Other |        310         633        5901         258        4507       11609

|   84.44069    526.1821        5901    507.9809    4589.396       11609

|

Non Hispanic Other |        101         317         214        3509        3959        8100

|   68.72013    428.2213    359.0829        3509    3734.976        8100

|

White non Hispanic |        615        5338        4403        5505      543276      559137

|   858.6274    5350.435    4486.582    5165.357      543276      559137

|

Total |      43641       24667       11542       10098      559873      649821

|      43641       24667       11542       10098      559873      649821

--------------------------------------------------------------------------------------------

* I generated the cell-by-cell pearson chisquare statistic, in order to figure out which cells fit the data worst by the quasi-symmetry model. Since the Pearson statistic is positive everywhere, it is more useful on a cell-by-cell basis than the LR chisquare statistic, which is positive in some cells and negative in others.

. gen pearson_quasi_indep=((count- quasi_indep_model)^2)/ quasi_indep_model

. table meth_num feth_num, contents(sum count sum quasi_indep_model sum pearson_quasi_indep ) row col cellwidth(10)

--------------------------------------------------------------------------------------------

husband's           |                         wife's race/ethnicity

race/ethnicity      | Black, non  Mexican Am  Hispanic O  Non Hispan  White non        Total

--------------------+-----------------------------------------------------------------------

Black, non Hispanic |      42521         291         412         393        2064       45681

|      42521    274.1622    229.8974    264.6786    2391.262       45681

|          0    1.034097    144.2441    62.21268    44.78817     252.279

|

Mexican American |         94       18088         612         433        6067       25294

|   108.2118       18088    565.4383    650.9836    5881.366       25294

|    1.86647           0    3.834181    72.99238    5.859167     84.5522

|

Hispanic Other |        310         633        5901         258        4507       11609

|   84.44069    526.1821        5901    507.9809    4589.396       11609

|   602.5176    21.68462           0    123.0173    1.479319    748.6989

|

Non Hispanic Other |        101         317         214        3509        3959        8100

|   68.72013    428.2213    359.0829        3509    3734.976        8100

|    15.1628    28.88735    58.61894           0    13.43702    116.1061

|

White non Hispanic |        615        5338        4403        5505      543276      559137

|   858.6274    5350.435    4486.582    5165.357      543276      559137

|   69.12699    .0288983     1.55706     22.3329           0    93.04585

|

Total |      43641       24667       11542       10098      559873      649821

|      43641       24667       11542       10098      559873      649821

|   688.6739    51.63497    208.2543    280.5553    65.56367    1294.682

--------------------------------------------------------------------------------------------

* Looking at this table, we see that cells 1,3 and 3,1 (black-other Hispanic intermarriage) contribute a lot to the total Pearson chisquare statistic of 1294.

. codebook meth_num

-----------------------------------------------------------------------------------------------------

meth_num                                                                     husband's race/ethnicity

-----------------------------------------------------------------------------------------------------

type:  numeric (byte)

label:  ethnicity

range:  [1,5]                        units:  1

unique values:  5                        missing .:  0/25

tabulation:  Freq.   Numeric  Label

5         1  Black, non Hispanic

5         2  Mexican American

5         3  Hispanic Other

5         4  Non Hispanic Other

5         5  White non Hispanic

* Now I generate the symmetric dummy variable to fit the two black-other Hispanic cells.

. gen byte black_othHisp=0

. replace black_othHisp=1 if (meth_num==1 & feth_num==3)|(feth_num==1 & meth_num==3)

. *this is the gender symmetric black-Other Hispanic interaction.

. *Which we picked because the black-other Hispanic cells are fit poorly by quasi-symmetry

. poisson count i.meth_num i.feth_num i.endogamy_diagonal_cat

Iteration 0:   log likelihood = -1622248.5

Iteration 1:   log likelihood = -451561.88  (backed up)

Iteration 2:   log likelihood = -374791.98  (backed up)

Iteration 3:   log likelihood = -199026.43  (backed up)

Iteration 4:   log likelihood =  -143118.6

Iteration 5:   log likelihood = -64630.014

Iteration 6:   log likelihood = -2062.1525

Iteration 7:   log likelihood = -650.23659

Iteration 8:   log likelihood = -648.54413

Iteration 9:   log likelihood = -648.54411

Poisson regression                                Number of obs   =         25

LR chi2(13)     = 3151666.22

Prob > chi2     =     0.0000

Log likelihood = -648.54411                       Pseudo R2       =     0.9996

---------------------------------------------------------------------------------------

count |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]

----------------------+----------------------------------------------------------------

meth_num |

Mexican American  |    .899968   .0213908    42.07   0.000     .8580427    .9418932

Hispanic Other  |   .6519274   .0222121    29.35   0.000     .6083924    .6954623

Non Hispanic Other  |   .4459202   .0231596    19.25   0.000     .4005282    .4913121

White non Hispanic  |   2.971213   .0235674   126.07   0.000     2.925022    3.017404

|

feth_num |

Mexican American  |   1.829598    .032363    56.53   0.000     1.766168    1.893028

Hispanic Other  |   1.653511   .0327399    50.50   0.000     1.589342     1.71768

Non Hispanic Other  |   1.794394   .0323429    55.48   0.000     1.731004    1.857785

White non Hispanic  |   3.995454   .0336881   118.60   0.000     3.929427    4.061482

|

endogamy_diagonal_cat |

1  |   6.873631   .0370393   185.58   0.000     6.801036    6.946227

2  |   3.289316   .0229991   143.02   0.000     3.244239    3.334393

3  |   2.593316   .0262895    98.64   0.000      2.54179    2.644843

4  |    2.13865   .0286843    74.56   0.000      2.08243     2.19487

5  |   2.454583   .0192917   127.24   0.000     2.416772    2.492394

|

_cons |   3.784122   .0367204   103.05   0.000     3.712151    3.856093

---------------------------------------------------------------------------------------

. poisgof

Deviance goodness-of-fit =  1067.088

Prob > chi2(11)          =    0.0000

Pearson goodness-of-fit  =  1294.682

Prob > chi2(11)          =    0.0000

. *That was quasi independenc

. poisson count i.meth_num i.feth_num i.endogamy_diagonal_cat i.black_othHisp

Iteration 0:   log likelihood = -1622150.1

Iteration 1:   log likelihood = -448544.31  (backed up)

Iteration 2:   log likelihood = -406174.34  (backed up)

Iteration 3:   log likelihood = -164842.09  (backed up)

Iteration 4:   log likelihood = -122114.39  (backed up)

Iteration 5:   log likelihood = -38616.448

Iteration 6:   log likelihood = -5553.1389

Iteration 7:   log likelihood = -541.68591

Iteration 8:   log likelihood = -407.86486

Iteration 9:   log likelihood = -407.61852

Iteration 10:  log likelihood = -407.61852

Poisson regression                                Number of obs   =         25

LR chi2(14)     = 3152148.07

Prob > chi2     =     0.0000

Log likelihood = -407.61852                       Pseudo R2       =     0.9997

---------------------------------------------------------------------------------------

count |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]

----------------------+----------------------------------------------------------------

meth_num |

Mexican American  |   1.011966    .022452    45.07   0.000     .9679614    1.055972

Hispanic Other  |   .7375092   .0229166    32.18   0.000     .6925934     .782425

Non Hispanic Other  |   .5583559   .0241463    23.12   0.000       .51103    .6056819

White non Hispanic  |   3.178408   .0259865   122.31   0.000     3.127476    3.229341

|

feth_num |

Mexican American  |   1.952846   .0332485    58.73   0.000      1.88768    2.018012

Hispanic Other  |   1.713306   .0330711    51.81   0.000     1.648488    1.778124

Non Hispanic Other  |   1.920179   .0332427    57.76   0.000     1.855024    1.985333

White non Hispanic  |   4.213423   .0356215   118.28   0.000     4.143607     4.28324

|

endogamy_diagonal_cat |

1  |   7.180968   .0406743   176.55   0.000     7.101248    7.260688

2  |   3.361407   .0238147   141.15   0.000     3.314731    3.408082

3  |   2.755277   .0277498    99.29   0.000     2.700889    2.809666

4  |   2.207766   .0293102    75.32   0.000     2.150319    2.265213

5  |   2.336756   .0206074   113.39   0.000     2.296366    2.377146

|

1.black_othHisp |   1.072106   .0447908    23.94   0.000     .9843172    1.159894

_cons |   3.476785   .0403842    86.09   0.000     3.397634    3.555937

---------------------------------------------------------------------------------------

* Among black married people who are not married to other black people, the log odds of being married to a person from Other-Hispanic background was significantly higher (increased the log odds by 1.07, increased the count by a factor of exp(1.07), or had a relative risk of exp(1.07)) compared to marriages to whites, non-Hispanics, and Mexican Americans. That is, among the non endogamous black married people, more than expected were married to people from the other Hispanic category.

. poisgof

Deviance goodness-of-fit =   585.237

Prob > chi2(10)          =    0.0000

Pearson goodness-of-fit  =  602.9136

Prob > chi2(10)          =    0.0000

* Adds one df, for black-other Hispanic interaction, and improves the fit by (1067-585=482) on 1 df, which is highly significant.

. predict QI_and_BOH

(option n assumed; predicted number of events)

. table meth_num feth_num, contents(sum count sum QI_and_BOH) row col cellwidth(10)

--------------------------------------------------------------------------------------------

husband's           |                         wife's race/ethnicity

race/ethnicity      | Black, non  Mexican Am  Hispanic O  Non Hispan  White non        Total

--------------------+-----------------------------------------------------------------------

Black, non Hispanic |      42521         291         412         393        2064       45681

|      42521    228.0651    524.3694    220.7353     2186.83       45681

|

Mexican American |         94       18088         612         433        6067       25294

|   89.01025       18088    493.7638    607.2439    6015.982       25294

|

Hispanic Other |        310         633        5901         258        4507       11609

|   197.6306    476.8205        5901    461.4959    4572.053       11609

|

Non Hispanic Other |        101         317         214        3509        3959        8100

|    56.5509    398.6114     313.703        3509    3822.135        8100

|

White non Hispanic |        615        5338        4403        5505      543276      559137

|   776.8082    5475.503    4309.164    5299.525      543276      559137

|

Total |      43641       24667       11542       10098      559873      649821

|      43641       24667       11542       10098      559873      649821

--------------------------------------------------------------------------------------------

. display 310+412

722

. display 197.63+524.37

722

*We have one term fitting the two cells for black-Other Hispanic intermarriage, so the sum of those two cells is the same in the actual and the predicted.

* Now add the gender specific black-other Hispanic term.

. gen byte husb_black_wife_OH=0

. replace husb_black_wife_OH=1 if meth_num==1 & feth_num==3

. table meth_num feth_num, contents(mean black_othHisp) cellwidth(10)

note: cellwidth too small, variable name truncated;

to increase cellwidth, specify cellwidth(#)

--------------------------------------------------------------------------------

husband's           |                   wife's race/ethnicity

race/ethnicity      | Black, non  Mexican Am  Hispanic O  Non Hispan  White non

--------------------+-----------------------------------------------------------

Black, non Hispanic |          0           0           1           0           0

Mexican American |          0           0           0           0           0

Hispanic Other |          1           0           0           0           0

Non Hispanic Other |          0           0           0           0           0

White non Hispanic |          0           0           0           0           0

--------------------------------------------------------------------------------

. table meth_num feth_num, contents(mean husb_black_wife_OH ) cellwidth(10)

note: cellwidth too small, variable name truncated;

to increase cellwidth, specify cellwidth(#)

--------------------------------------------------------------------------------

husband's           |                   wife's race/ethnicity

race/ethnicity      | Black, non  Mexican Am  Hispanic O  Non Hispan  White non

--------------------+-----------------------------------------------------------

Black, non Hispanic |          0           0           1           0           0

Mexican American |          0           0           0           0           0

Hispanic Other |          0           0           0           0           0

Non Hispanic Other |          0           0           0           0           0

White non Hispanic |          0           0           0           0           0

--------------------------------------------------------------------------------

. poisson count i.meth_num i.feth_num i.endogamy_diagonal_cat i.black_othHisp i.husb_black_wife_OH

Iteration 0:   log likelihood = -1622147.3

Iteration 1:   log likelihood = -448422.28  (backed up)

Iteration 2:   log likelihood = -405890.17  (backed up)

Iteration 3:   log likelihood = -164210.39  (backed up)

Iteration 4:   log likelihood = -121629.82  (backed up)

Iteration 5:   log likelihood = -38857.644

Iteration 6:   log likelihood = -5529.6282

Iteration 7:   log likelihood = -491.60417

Iteration 8:   log likelihood = -355.79505

Iteration 9:   log likelihood = -355.54662

Iteration 10:  log likelihood = -355.54662

Poisson regression                                Number of obs   =         25

LR chi2(15)     = 3152252.22

Prob > chi2     =     0.0000

Log likelihood = -355.54662                       Pseudo R2       =     0.9998

---------------------------------------------------------------------------------------

count |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]

----------------------+----------------------------------------------------------------

meth_num |

Mexican American  |   .9702156   .0224704    43.18   0.000     .9261743    1.014257

Hispanic Other  |   .6751551   .0234341    28.81   0.000     .6292252     .721085

Non Hispanic Other  |   .5166051   .0241635    21.38   0.000     .4692455    .5639646

White non Hispanic  |   3.136658   .0260024   120.63   0.000     3.085694    3.187621

|

feth_num |

Mexican American  |   2.082757   .0372736    55.88   0.000     2.009703    2.155812

Hispanic Other  |   1.864941   .0377599    49.39   0.000     1.790933    1.938949

Non Hispanic Other  |    2.05009   .0372684    55.01   0.000     1.977046    2.123135

White non Hispanic  |   4.343335   .0394049   110.22   0.000     4.266103    4.420567

|

endogamy_diagonal_cat |

1  |   7.269129   .0429614   169.20   0.000     7.184926    7.353332

2  |   3.361407   .0238147   141.15   0.000     3.314731    3.408082

3  |   2.754156   .0277426    99.28   0.000     2.699782    2.808531

4  |   2.207766   .0293102    75.32   0.000     2.150319    2.265213

5  |   2.336756   .0206074   113.39   0.000     2.296366    2.377146

|

1.black_othHisp |   1.672793   .0697805    23.97   0.000     1.536025     1.80956

1.husb_black_wife_OH |  -.9053349   .0873382   -10.37   0.000    -1.076515   -.7341552

_cons |   3.388624   .0426868    79.38   0.000      3.30496    3.472289

---------------------------------------------------------------------------------------

. poisgof

Deviance goodness-of-fit =  481.0932

Prob > chi2(9)           =    0.0000

Pearson goodness-of-fit  =   494.695

Prob > chi2(9)           =    0.0000

. *Now let's see what black-other Hispanic intermarriage looks like of we don't mark out the endogamy

>  diagonal.

. poisson count i.meth_num i.feth_num i.black_othHisp

Iteration 0:   log likelihood = -312574.59

Iteration 1:   log likelihood = -232297.17

Iteration 2:   log likelihood = -224728.61

Iteration 3:   log likelihood = -224682.36

Iteration 4:   log likelihood = -224682.36

Poisson regression                                Number of obs   =         25

LR chi2(9)      = 2703598.60

Prob > chi2     =     0.0000

Log likelihood = -224682.36                       Pseudo R2       =     0.8575

-------------------------------------------------------------------------------------

count |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]

--------------------+----------------------------------------------------------------

meth_num |

Mexican American  |  -.6016685   .0078446   -76.70   0.000    -.6170436   -.5862934

Hispanic Other  |  -1.341195   .0104305  -128.58   0.000    -1.361639   -1.320752

Non Hispanic Other  |  -1.740372   .0120606  -144.30   0.000     -1.76401   -1.716733

White non Hispanic  |   2.494159   .0048776   511.35   0.000     2.484599    2.503719

|

feth_num |

Mexican American  |  -.5811256   .0079728   -72.89   0.000    -.5967521   -.5654992

Hispanic Other  |  -1.299468    .010508  -123.66   0.000    -1.320063   -1.278872

Non Hispanic Other  |  -1.474255   .0110479  -133.44   0.000    -1.495908   -1.452601

White non Hispanic  |   2.541118   .0049812   510.15   0.000     2.531355    2.550881

|

1.black_othHisp |  -.8394343   .0379587   -22.11   0.000    -.9138319   -.7650366

_cons |   8.048426   .0066011  1219.26   0.000     8.035488    8.061364

-------------------------------------------------------------------------------------

*Although the black-Other Hispanic association was positive and significant compared to other forms of non endogamy for black spouses, that is compared to marriage to other non-black groups, when we get rid of the terms for the endogamy diagonal, the black-other Hispanic association is strongly negative, because now we are also comparing to black-black marriage, and that is always more common. Also, without the endogamy diagonal, the model will fit poorly.

. poisgof

Deviance goodness-of-fit =  449134.7

Prob > chi2(15)          =    0.0000

Pearson goodness-of-fit  =   1153768

Prob > chi2(15)          =    0.0000

. *when not controlling for the endogamy diagonal, models tend to fit poorly

. gen QS=0

. replace QS= (meth_num*10+ feth_num) if feth_num>meth_num

. replace QS= (feth_num*10+ meth_num) if meth_num>feth_num

*Constructing a set of numbers that are symmetric and otherwise different in every cell.

. table meth_num feth_num, contents(mean QS) cellwidth(10)

--------------------------------------------------------------------------------

husband's           |                   wife's race/ethnicity

race/ethnicity      | Black, non  Mexican Am  Hispanic O  Non Hispan  White non

--------------------+-----------------------------------------------------------

Black, non Hispanic |          0          12          13          14          15

Mexican American |         12           0          23          24          25

Hispanic Other |         13          23           0          34          35

Non Hispanic Other |         14          24          34           0          45

White non Hispanic |         15          25          35          45           0

--------------------------------------------------------------------------------

* Quasi symmetry treats the terms as

. poisson count i.meth_num i.feth_num i.QS

Iteration 0:   log likelihood =   -1621737

Iteration 1:   log likelihood = -438727.68  (backed up)

Iteration 2:   log likelihood =  -371941.3  (backed up)

Iteration 3:   log likelihood = -207096.11  (backed up)

Iteration 4:   log likelihood = -185110.43

Iteration 5:   log likelihood = -90707.459

Iteration 6:   log likelihood =  -16934.19

Iteration 7:   log likelihood = -573.01331

Iteration 8:   log likelihood = -173.51256

Iteration 9:   log likelihood =  -170.2886

Iteration 10:  log likelihood = -170.28779

Iteration 11:  log likelihood = -170.28779

Poisson regression                                Number of obs   =         25

LR chi2(18)     = 3152622.73

Prob > chi2     =     0.0000

Log likelihood = -170.28779                       Pseudo R2       =     0.9999

-------------------------------------------------------------------------------------

count |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]

--------------------+----------------------------------------------------------------

meth_num |

Mexican American  |  -.8865297   .0195915   -45.25   0.000    -.9249284   -.8481311

Hispanic Other  |  -1.464572   .0203211   -72.07   0.000    -1.504401   -1.424743

Non Hispanic Other  |  -1.916449   .0213965   -89.57   0.000    -1.958386   -1.874513

White non Hispanic  |   .7607015   .0177297    42.91   0.000     .7259519    .7954511

|

feth_num |

Mexican American  |   .0317804   .0195915     1.62   0.105    -.0066182    .0701791

Hispanic Other  |  -.5103041   .0203211   -25.11   0.000    -.5501328   -.4704754

Non Hispanic Other  |  -.5782177   .0213965   -27.02   0.000    -.6201542   -.5362813

White non Hispanic  |   1.786918   .0177297   100.79   0.000     1.752168    1.821668

|

QS |

12  |  -5.072186   .0518097   -97.90   0.000    -5.173731   -4.970641

13  |  -3.891192   .0387965  -100.30   0.000    -3.967232   -3.815152

14  |  -4.109943    .047238   -87.01   0.000    -4.202528   -4.017358

15  |  -4.857751    .021174  -229.42   0.000    -4.899251    -4.81625

23  |  -2.809359   .0293163   -95.83   0.000    -2.866818     -2.7519

24  |  -3.078001    .037751   -81.53   0.000    -3.151991    -3.00401

25  |  -2.856983   .0101084  -282.64   0.000    -2.876795   -2.837171

34  |  -2.977466   .0473135   -62.93   0.000    -3.070199   -2.884733

35  |   -2.54299    .012457  -204.14   0.000    -2.567405   -2.518574

45  |  -2.234246   .0134023  -166.71   0.000    -2.260515   -2.207978

|

_cons |   10.65775   .0048495  2197.69   0.000     10.64825    10.66726

-------------------------------------------------------------------------------------

. poisgof

Deviance goodness-of-fit =  110.5756

Prob > chi2(6)           =    0.0000

Pearson goodness-of-fit  =  117.1273

Prob > chi2(6)           =    0.0000

. poisson count i.meth_num i.feth_num i.endogamy_diagonal_cat i.QS

* Even though it is not obvious why this must be so, the QS terms are the full set of symmetric interaction terms that can be included, so adding the endogamy diagonal terms (which are also symmetric with respect to husband and wife) yield the same model, but with a different set of terms displayed:

note: 15.QS omitted because of collinearity

note: 25.QS omitted because of collinearity

note: 34.QS omitted because of collinearity

note: 35.QS omitted because of collinearity

note: 45.QS omitted because of collinearity

Iteration 0:   log likelihood =   -1621737

Iteration 1:   log likelihood = -438727.66  (backed up)

Iteration 2:   log likelihood = -371941.28  (backed up)

Iteration 3:   log likelihood = -207096.16  (backed up)

Iteration 4:   log likelihood = -185110.46

Iteration 5:   log likelihood = -90707.453

Iteration 6:   log likelihood = -16934.185

Iteration 7:   log likelihood = -573.01339

Iteration 8:   log likelihood = -173.51256

Iteration 9:   log likelihood =  -170.2886

Iteration 10:  log likelihood = -170.28779

Iteration 11:  log likelihood = -170.28779

Poisson regression                                Number of obs   =         25

LR chi2(18)     = 3152622.73

Prob > chi2     =     0.0000

Log likelihood = -170.28779                       Pseudo R2       =     0.9999

---------------------------------------------------------------------------------------

count |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]

----------------------+----------------------------------------------------------------

meth_num |

Mexican American  |   1.114238   .0245816    45.33   0.000     1.066059    1.162417

Hispanic Other  |    .850189    .025271    33.64   0.000     .8006588    .8997193

Non Hispanic Other  |    .707055   .0260209    27.17   0.000     .6560549    .7580551

White non Hispanic  |   3.818683   .0528711    72.23   0.000     3.715057    3.922308

|

feth_num |

Mexican American  |   2.032548   .0343926    59.10   0.000      1.96514    2.099956

Hispanic Other  |   1.804457   .0346164    52.13   0.000      1.73661    1.872304

Non Hispanic Other  |   2.045286   .0344563    59.36   0.000     1.977753     2.11282

White non Hispanic  |   4.844899    .058194    83.25   0.000     4.730841    4.958957

|

endogamy_diagonal_cat |

1  |   7.915732   .0643011   123.10   0.000     7.789704     8.04176

2  |   3.914197   .0524013    74.70   0.000     3.811492    4.016901

3  |   3.286209   .0501069    65.58   0.000     3.188002    3.384417

4  |   2.668723   .0513256    52.00   0.000     2.568127     2.76932

5  |    1.79977   .0483798    37.20   0.000     1.704947    1.894592

|

QS |

12  |   .8427781   .0735992    11.45   0.000     .6985263      .98703

13  |   1.709778   .0634045    26.97   0.000     1.585508    1.834049

14  |   1.182285   .0681776    17.34   0.000     1.048659     1.31591

15  |          0  (omitted)

23  |   .7908445   .0558517    14.16   0.000     .6813773    .9003117

24  |   .2134592   .0604599     3.53   0.000     .0949599    .3319584

25  |          0  (omitted)

34  |          0  (omitted)

35  |          0  (omitted)

45  |          0  (omitted)

|

_cons |   2.742022    .064118    42.77   0.000     2.616353    2.867691

---------------------------------------------------------------------------------------

. poisgof

Deviance goodness-of-fit =  110.5756

Prob > chi2(6)           =    0.0000

Pearson goodness-of-fit  =  117.1273

Prob > chi2(6)           =    0.0000

. *This is another version of the QS, quasi symmetry model, which has the endogamy diagonal displa

> cing 5 of the off-diagonal QS terms.

. *gen ID_QS=(50/649821)*(abs(count- constant_only_class ))

. predict QS_predicted

(option n assumed; predicted number of events)

. *gen ID_QS=(50/649821)*(abs(count- QS_predicted ))

. gen ID_QS=(50/649821)*(abs(count- QS_predicted ))

. table meth_num feth_num, contents(sum ID_QS) row col cellwidth(10)

--------------------------------------------------------------------------------------------

husband's           |                         wife's race/ethnicity

race/ethnicity      | Black, non  Mexican Am  Hispanic O  Non Hispan  White non        Total

--------------------+-----------------------------------------------------------------------

Black, non Hispanic |          0    .0012189    .0084073    .0001272    .0070611    .0168145

Mexican American |   .0012189           0     .001669     .001508    .0043959    .0087917

Hispanic Other |   .0084073     .001669           0    .0017512     .008325    .0201525

Non Hispanic Other |   .0001272     .001508    .0017512           0    .0031319    .0065183

White non Hispanic |   .0070611    .0043959     .008325    .0031319           0    .0229139

|

Total |   .0168145    .0087917    .0201525    .0065183    .0229139     .075191

--------------------------------------------------------------------------------------------

. poisson count i.meth_num i.feth_num i.endogamy_diagonal_cat i.black_othHisp i.husb_black_wife_OH

Iteration 0:   log likelihood = -1622147.3

Iteration 1:   log likelihood = -448422.28  (backed up)

Iteration 2:   log likelihood = -405890.17  (backed up)

Iteration 3:   log likelihood = -164210.39  (backed up)

Iteration 4:   log likelihood = -121629.82  (backed up)

Iteration 5:   log likelihood = -38857.644

Iteration 6:   log likelihood = -5529.6282

Iteration 7:   log likelihood = -491.60417

Iteration 8:   log likelihood = -355.79505

Iteration 9:   log likelihood = -355.54662

Iteration 10:  log likelihood = -355.54662

Poisson regression                                Number of obs   =         25

LR chi2(15)     = 3152252.22

Prob > chi2     =     0.0000

Log likelihood = -355.54662                       Pseudo R2       =     0.9998

---------------------------------------------------------------------------------------

count |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]

----------------------+----------------------------------------------------------------

meth_num |

Mexican American  |   .9702156   .0224704    43.18   0.000     .9261743    1.014257

Hispanic Other  |   .6751551   .0234341    28.81   0.000     .6292252     .721085

Non Hispanic Other  |   .5166051   .0241635    21.38   0.000     .4692455    .5639646

White non Hispanic  |   3.136658   .0260024   120.63   0.000     3.085694    3.187621

|

feth_num |

Mexican American  |   2.082757   .0372736    55.88   0.000     2.009703    2.155812

Hispanic Other  |   1.864941   .0377599    49.39   0.000     1.790933    1.938949

Non Hispanic Other  |    2.05009   .0372684    55.01   0.000     1.977046    2.123135

White non Hispanic  |   4.343335   .0394049   110.22   0.000     4.266103    4.420567

|

endogamy_diagonal_cat |

1  |   7.269129   .0429614   169.20   0.000     7.184926    7.353332

2  |   3.361407   .0238147   141.15   0.000     3.314731    3.408082

3  |   2.754156   .0277426    99.28   0.000     2.699782    2.808531

4  |   2.207766   .0293102    75.32   0.000     2.150319    2.265213

5  |   2.336756   .0206074   113.39   0.000     2.296366    2.377146

|

1.black_othHisp |   1.672793   .0697805    23.97   0.000     1.536025     1.80956

1.husb_black_wife_OH |  -.9053349   .0873382   -10.37   0.000    -1.076515   -.7341552

_cons |   3.388624   .0426868    79.38   0.000      3.30496    3.472289

---------------------------------------------------------------------------------------

. * lincom can test whether two terms, for instance two endogamy terms, are significantly different from each other. To test whether the force of endogamy is the same across groups or different, that would be a goodness of fit test between the two models, one which had a single term for endogamy, and the Quasi-Independence model with 5 endogamy terms.

.

. lincom 4.endogamy_diagonal_cat-5.endogamy_diagonal_cat

( 1)  [count]4.endogamy_diagonal_cat - [count]5.endogamy_diagonal_cat = 0

------------------------------------------------------------------------------

count |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]

-------------+----------------------------------------------------------------

(1) |  -.1289895   .0431923    -2.99   0.003    -.2136449   -.0443342

------------------------------------------------------------------------------

*How do we know that the goodness of fit test at the top of the model is compared to the constant only model? We run the constant only model and compute the -2 times log likelihood of the difference of the two models.

. display (-1576481-(-355.54))*-2

3152250.9

. log close

name:  <unnamed>

log:  C:\Users\mexmi\Documents\newer web pages\Soc_382\logs\2nd loglin intermar class.log

log type:  text

closed on:  31 Jan 2019, 14:04:47

--------------------------------------------------------------------------------------------------