import numpy as np

class Bunch(dict):
    def __init__(self, **kw):
        dict.__init__(self, kw)
        self.__dict__  = self

        for i,att in enumerate(['params', 'bse', 'tvalues', 'pvalues']):
            self[att] = self.params_table[:,i]


est = dict(
           rank = 7,
           N = 10,
           ic = 3,
           k = 8,
           k_eq = 1,
           k_dv = 1,
           converged = 1,
           rc = 0,
           k_autoCns = 1,
           ll = -33.45804471711131,
           k_eq_model = 1,
           ll_0 = -349.6684656479622,
           df_m = 6,
           chi2 = 632.4208418617018,
           p = 2.3617193197e-133,
           r2_p = .9043149497192691,
           cmdline = "poisson deaths lnpyears smokes i.agecat",
           cmd = "poisson",
           predict = "poisso_p",
           estat_cmd = "poisson_estat",
           gof = "poiss_g",
           chi2type = "LR",
           opt = "moptimize",
           vce = "oim",
           title = "Poisson regression",
           user = "poiss_lf",
           crittype = "log likelihood",
           ml_method = "e2",
           singularHmethod = "m-marquardt",
           technique = "nr",
           which = "max",
           depvar = "deaths",
           properties = "b V",
          )

params_table = np.array([
     .66308184237808,  .63593388706566,  1.0426899019923,  .29709193621918,
    -.58332567281917,  1.9094893575753, np.nan,  1.9599639845401,
                   0,  .84966723812924,  .94279599903649,  .90122066597395,
     .36747100512904, -.99817896475073,  2.6975134410092, np.nan,
     1.9599639845401,                0,                0, np.nan,
    np.nan, np.nan, np.nan, np.nan,
    np.nan,  1.9599639845401,                0,  1.3944392032504,
     .25613243411925,  5.4442117338454,  5.203529593e-08,   .8924288571041,
     1.8964495493967, np.nan,  1.9599639845401,                0,
      2.389284381366,  .48305517266329,  4.9461935542328,  7.567871319e-07,
     1.4425136404002,  3.3360551223318, np.nan,  1.9599639845401,
                   0,  2.8385093615484,  .98099727008295,  2.8934936397003,
     .00380982006764,  .91579004325369,  4.7612286798431, np.nan,
     1.9599639845401,                0,  2.9103531988515,   1.500316321385,
     1.9398263935201,  .05240079188831, -.03021275648066,  5.8509191541838,
    np.nan,  1.9599639845401,                0,  -4.724924181641,
     6.0276019460727, -.78388125558284,  .43310978942119, -16.538806909087,
      7.088958545805, np.nan,  1.9599639845401,                0
    ]).reshape(8,9)

params_table_colnames = 'b se z pvalue ll ul df crit eform'.split()

params_table_rownames = 'lnpyears smokes 1b.agecat 2.agecat 3.agecat 4.agecat 5.agecat _cons'.split()

cov = np.array([
     .40441190871844, -.59566294916097,                0,   .1055698685775,
     .28413388045122,  .61269322798077,  .94624135329227, -3.8311942353131,
    -.59566294916097,  .88886429579921,                0, -.15587944298625,
     -.4190789999425, -.90299843943229, -1.3940094688194,  5.6335527795822,
                   0,                0,                0,                0,
                   0,                0,                0,                0,
      .1055698685775, -.15587944298625,                0,  .06560382380785,
     .10360281461667,  .18937107288073,  .27643306166968,  -1.029211453947,
     .28413388045122,  -.4190789999425,                0,  .10360281461667,
     .23334229983676,  .45990880867889,  .69424104947043, -2.7206801001387,
     .61269322798077, -.90299843943229,                0,  .18937107288073,
     .45990880867889,  .96235564391021,  1.4630024143274, -5.8333014154113,
     .94624135329227, -1.3940094688194,                0,  .27643306166968,
     .69424104947043,  1.4630024143274,  2.2509490642142,  -8.993394678922,
    -3.8311942353131,  5.6335527795822,                0,  -1.029211453947,
    -2.7206801001387, -5.8333014154113,  -8.993394678922,  36.331985220299
    ]).reshape(8,8)

cov_colnames = 'lnpyears smokes 1b.agecat 2.agecat 3.agecat 4.agecat 5.agecat _cons'.split()

cov_rownames = 'lnpyears smokes 1b.agecat 2.agecat 3.agecat 4.agecat 5.agecat _cons'.split()



results_noexposure_noconstraint = Bunch(
                params_table=params_table,
                params_table_colnames=params_table_colnames,
                params_table_rownames=params_table_rownames,
                cov=cov,
                cov_colnames=cov_colnames,
                cov_rownames=cov_rownames,
                **est
                )

est = dict(
           rank = 6,
           N = 10,
           ic = 3,
           k = 7,
           k_eq = 1,
           k_dv = 1,
           converged = 1,
           rc = 0,
           k_autoCns = 1,
           ll = -33.6001534405213,
           k_eq_model = 1,
           ll_0 = -495.0676356770329,
           df_m = 5,
           chi2 = 922.9349644730232,
           p = 2.8920463572e-197,
           r2_p = .9321301757191799,
           cmdline = "poisson deaths smokes i.agecat, exposure(pyears)",
           cmd = "poisson",
           predict = "poisso_p",
           estat_cmd = "poisson_estat",
           offset = "ln(pyears)",
           gof = "poiss_g",
           chi2type = "LR",
           opt = "moptimize",
           vce = "oim",
           title = "Poisson regression",
           user = "poiss_lf",
           crittype = "log likelihood",
           ml_method = "e2",
           singularHmethod = "m-marquardt",
           technique = "nr",
           which = "max",
           depvar = "deaths",
           properties = "b V",
          )

params_table = np.array([
     .35453563725291,  .10737411818853,  3.3018723993653,  .00096041750265,
     .14408623273163,  .56498504177418, np.nan,  1.9599639845401,
                   0,                0, np.nan, np.nan,
    np.nan, np.nan, np.nan, np.nan,
     1.9599639845401,                0,  1.4840070063099,  .19510337263434,
      7.606260139291,  2.821411159e-14,  1.1016114226842,  1.8664025899355,
    np.nan,  1.9599639845401,                0,  2.6275051184579,
     .18372726944827,  14.301116684248,  2.153264398e-46,  2.2674062873614,
     2.9876039495544, np.nan,  1.9599639845401,                0,
      3.350492785161,  .18479918093323,  18.130452571495,  1.832448146e-73,
     2.9882930461593,  3.7126925241626, np.nan,  1.9599639845401,
                   0,  3.7000964518246,  .19221951212105,   19.24932807807,
     1.430055953e-82,  3.3233531309415,  4.0768397727077, np.nan,
     1.9599639845401,                0,  -7.919325711822,  .19176181876223,
    -41.297719029467,                0, -8.2951719702059, -7.5434794534381,
    np.nan,  1.9599639845401,                0]).reshape(7,9)

params_table_colnames = 'b se z pvalue ll ul df crit eform'.split()

params_table_rownames = 'smokes 1b.agecat 2.agecat 3.agecat 4.agecat 5.agecat _cons'.split()

cov = np.array([
     .01152920125677,                0, -.00061561668833, -.00090117889461,
    -.00087280941113, -.00045274641397, -.00921219275997,                0,
                   0,                0,                0,                0,
                   0,                0, -.00061561668833,                0,
      .0380653260133,  .02945988432334,  .02945836949789,   .0294359396881,
     -.0289198676971, -.00090117889461,                0,  .02945988432334,
     .03375570953892,   .0294799877675,  .02944715358419, -.02869169455392,
    -.00087280941113,                0,  .02945836949789,   .0294799877675,
     .03415073727359,  .02944603952766, -.02871436265941, -.00045274641397,
                   0,   .0294359396881,  .02944715358419,  .02944603952766,
     .03694834084006, -.02905000614546, -.00921219275997,                0,
     -.0289198676971, -.02869169455392, -.02871436265941, -.02905000614546,
       .036772595135]).reshape(7,7)

cov_colnames = 'smokes 1b.agecat 2.agecat 3.agecat 4.agecat 5.agecat _cons'.split()

cov_rownames = 'smokes 1b.agecat 2.agecat 3.agecat 4.agecat 5.agecat _cons'.split()


results_exposure_noconstraint = Bunch(
                params_table=params_table,
                params_table_colnames=params_table_colnames,
                params_table_rownames=params_table_rownames,
                cov=cov,
                cov_colnames=cov_colnames,
                cov_rownames=cov_rownames,
                **est
                )

est = dict(
           rank = 6,
           N = 10,
           ic = 4,
           k = 8,
           k_eq = 1,
           k_dv = 1,
           converged = 1,
           rc = 0,
           k_autoCns = 1,
           ll = -33.46699798755848,
           k_eq_model = 1,
           df_m = 5,
           chi2 = 452.5895246742914,
           p = 1.35732711092e-95,
           r2_p = np.nan,
           cmdline = "poisson deaths lnpyears smokes i.agecat, constraints(1)",
           cmd = "poisson",
           predict = "poisso_p",
           estat_cmd = "poisson_estat",
           gof = "poiss_g",
           chi2type = "Wald",
           opt = "moptimize",
           vce = "oim",
           title = "Poisson regression",
           user = "poiss_lf",
           crittype = "log likelihood",
           ml_method = "e2",
           singularHmethod = "m-marquardt",
           technique = "nr",
           which = "max",
           depvar = "deaths",
           properties = "b V",
          )

params_table = np.array([
     .57966535352347,  .13107152221057,  4.4225117992619,  9.756001957e-06,
     .32276989059191,  .83656081645503, np.nan,  1.9599639845401,
                   0,  .97254074124891,  .22289894431919,  4.3631464663029,
     .00001282050472,   .5356668381913,  1.4094146443065, np.nan,
     1.9599639845401,                0,                0, np.nan,
    np.nan, np.nan, np.nan, np.nan,
    np.nan,  1.9599639845401,                0,  1.3727621378494,
     .19798042377276,  6.9338276567436,  4.096036246e-12,  .98472763761078,
      1.760796638088, np.nan,  1.9599639845401,                0,
     2.3307703209845,  .20530981936838,     11.352454199,  7.210981748e-30,
       1.92837046935,  2.7331701726189, np.nan,  1.9599639845401,
                   0,    2.71338890728,  .29962471107816,  9.0559583604312,
     1.353737255e-19,  2.1261352646886,  3.3006425498714, np.nan,
     1.9599639845401,                0,    2.71338890728,  .29962471107816,
     9.0559583604312,  1.353737255e-19,  2.1261352646886,  3.3006425498714,
    np.nan,  1.9599639845401,                0, -3.9347864312059,
     1.2543868840549, -3.1368204508696,  .00170790683415, -6.3933395466329,
     -1.476233315779, np.nan,  1.9599639845401,                0
    ]).reshape(8,9)

params_table_colnames = 'b se z pvalue ll ul df crit eform'.split()

params_table_rownames = 'lnpyears smokes 1b.agecat 2.agecat 3.agecat 4.agecat 5.agecat _cons'.split()

cov = np.array([
      .0171797439346, -.02561346650005,                0,  .00445310785396,
     .01204526460873,  .03142116278001,  .03142116278001, -.16245493266167,
    -.02561346650005,  .04968393937861,                0,  -.0069699735991,
    -.01845598801461, -.04723465558226, -.04723465558226,   .2326939064726,
                   0,                0,                0,                0,
                   0,                0,                0,                0,
     .00445310785396,  -.0069699735991,                0,  .03919624819724,
     .03254829669461,  .03756752462584,  .03756752462584, -.07124751761252,
     .01204526460873, -.01845598801461,                0,  .03254829669461,
     .04215212192908,  .05145895528528,  .05145895528528, -.14290240509701,
     .03142116278001, -.04723465558226,                0,  .03756752462584,
     .05145895528528,  .08977496748867,  .08977496748867, -.32621483141938,
     .03142116278001, -.04723465558226,                0,  .03756752462584,
     .05145895528528,  .08977496748867,  .08977496748867, -.32621483141938,
    -.16245493266167,   .2326939064726,                0, -.07124751761252,
    -.14290240509701, -.32621483141938, -.32621483141938,  1.5734864548889
    ]).reshape(8,8)

cov_colnames = 'lnpyears smokes 1b.agecat 2.agecat 3.agecat 4.agecat 5.agecat _cons'.split()

cov_rownames = 'lnpyears smokes 1b.agecat 2.agecat 3.agecat 4.agecat 5.agecat _cons'.split()


results_noexposure_constraint = Bunch(
                params_table=params_table,
                params_table_colnames=params_table_colnames,
                params_table_rownames=params_table_rownames,
                cov=cov,
                cov_colnames=cov_colnames,
                cov_rownames=cov_rownames,
                **est
                )

est = dict(
           rank = 5,
           N = 10,
           ic = 3,
           k = 7,
           k_eq = 1,
           k_dv = 1,
           converged = 1,
           rc = 0,
           k_autoCns = 1,
           ll = -38.45090497564205,
           k_eq_model = 1,
           df_m = 4,
           chi2 = 641.6446542589836,
           p = 1.5005477751e-137,
           r2_p = np.nan,
           cmdline = "poisson deaths smokes i.agecat, exposure(pyears) constraints(1)",
           cmd = "poisson",
           predict = "poisso_p",
           estat_cmd = "poisson_estat",
           offset = "ln(pyears)",
           gof = "poiss_g",
           chi2type = "Wald",
           opt = "moptimize",
           vce = "oim",
           title = "Poisson regression",
           user = "poiss_lf",
           crittype = "log likelihood",
           ml_method = "e2",
           singularHmethod = "m-marquardt",
           technique = "nr",
           which = "max",
           depvar = "deaths",
           properties = "b V",
          )

params_table = np.array([
     .34304077058284,   .1073083520206,   3.196776058186,  .00138972774083,
     .13272026538212,  .55336127578356, np.nan,  1.9599639845401,
                   0,                0, np.nan, np.nan,
    np.nan, np.nan, np.nan, np.nan,
     1.9599639845401,                0,  1.4846230896448,  .19510453584194,
     7.6093724999174,  2.754298692e-14,  1.1022252261742,  1.8670209531154,
    np.nan,  1.9599639845401,                0,  2.6284071093765,
     .18373002757074,  14.305811326156,  2.012766793e-46,  2.2683028724593,
     2.9885113462937, np.nan,  1.9599639845401,                0,
     3.4712405808805,  .17983994458502,  19.301833020969,  5.183735658e-83,
     3.1187607665121,  3.8237203952488, np.nan,  1.9599639845401,
                   0,  3.4712405808805,  .17983994458502,  19.301833020969,
     5.183735658e-83,  3.1187607665121,  3.8237203952488, np.nan,
     1.9599639845401,                0, -7.9101515866812,  .19164951521841,
    -41.274049546467,                0, -8.2857777341639, -7.5345254391986,
    np.nan,  1.9599639845401,                0]).reshape(7,9)

params_table_colnames = 'b se z pvalue ll ul df crit eform'.split()

params_table_rownames = 'smokes 1b.agecat 2.agecat 3.agecat 4.agecat 5.agecat _cons'.split()

cov = np.array([
     .01151508241338,                0, -.00061947268694, -.00090708285562,
    -.00074959767622, -.00074959767622, -.00917958318314,                0,
                   0,                0,                0,                0,
                   0,                0, -.00061947268694,                0,
      .0380657799061,  .02946056271023,   .0294520905375,   .0294520905375,
    -.02891793401778, -.00090708285562,                0,  .02946056271023,
     .03375672303114,  .02947081310555,  .02947081310555, -.02868865719866,
    -.00074959767622,                0,   .0294520905375,  .02947081310555,
     .03234240566834,  .03234240566834, -.02881420109427, -.00074959767622,
                   0,   .0294520905375,  .02947081310555,  .03234240566834,
     .03234240566834, -.02881420109427, -.00917958318314,                0,
    -.02891793401778, -.02868865719866, -.02881420109427, -.02881420109427,
     .03672953668345]).reshape(7,7)

cov_colnames = 'smokes 1b.agecat 2.agecat 3.agecat 4.agecat 5.agecat _cons'.split()

cov_rownames = 'smokes 1b.agecat 2.agecat 3.agecat 4.agecat 5.agecat _cons'.split()


results_exposure_constraint = Bunch(
                params_table=params_table,
                params_table_colnames=params_table_colnames,
                params_table_rownames=params_table_rownames,
                cov=cov,
                cov_colnames=cov_colnames,
                cov_rownames=cov_rownames,
                **est
                )

est = dict(
           rank = 6,
           N = 10,
           ic = 3,
           k = 8,
           k_eq = 1,
           k_dv = 1,
           converged = 1,
           rc = 0,
           k_autoCns = 1,
           ll = -33.78306559091298,
           k_eq_model = 1,
           df_m = 5,
           chi2 = 526.719430888018,
           p = 1.3614066522e-111,
           r2_p = np.nan,
           cmdline = "poisson deaths lnpyears smokes i.agecat, constraints(2)",
           cmd = "poisson",
           predict = "poisso_p",
           estat_cmd = "poisson_estat",
           gof = "poiss_g",
           chi2type = "Wald",
           opt = "moptimize",
           vce = "oim",
           title = "Poisson regression",
           user = "poiss_lf",
           crittype = "log likelihood",
           ml_method = "e2",
           singularHmethod = "m-marquardt",
           technique = "nr",
           which = "max",
           depvar = "deaths",
           properties = "b V",
          )

params_table = np.array([
     1.1598786864273,  .13082965708054,  8.8655639119598,  7.611783820e-19,
     .90345727043975,  1.4163001024149, np.nan,  1.9599639845401,
                   0,  .12111539473831,  .22317899375276,  .54268277090847,
     .58734823873758, -.31630739512299,  .55853818459962, np.nan,
     1.9599639845401,                0,                0, np.nan,
    np.nan, np.nan, np.nan, np.nan,
    np.nan,  1.9599639845401,                0,  1.5276244194375,
     .19848759770871,  7.6963217705896,  1.400389019e-14,  1.1385958765506,
     1.9166529623245, np.nan,  1.9599639845401,                0,
     2.7415571106656,  .20647039325801,  13.278209371354,  3.097119459e-40,
     2.3368825760061,  3.1462316453252, np.nan,  1.9599639845401,
                   0,   3.587300073596,  .30160673316211,  11.893965482753,
     1.272196529e-32,  2.9961617391034,  4.1784384080885, np.nan,
     1.9599639845401,                0,   4.087300073596,  .30160673316211,
     13.551753406643,  7.735990122e-42,  3.4961617391034,  4.6784384080885,
    np.nan,  1.9599639845401,                0, -9.4376201542802,
     1.2537557101599, -7.5274792990385,  5.172920628e-14, -11.894936191605,
    -6.9803041169553, np.nan,  1.9599639845401,                0
    ]).reshape(8,9)

params_table_colnames = 'b se z pvalue ll ul df crit eform'.split()

params_table_rownames = 'lnpyears smokes 1b.agecat 2.agecat 3.agecat 4.agecat 5.agecat _cons'.split()

cov = np.array([
     .01711639917181, -.02559852137367,                0,  .00475026273828,
       .012305588195,  .03167368550108,  .03167368550108, -.16210959536359,
    -.02559852137367,   .0498088632525,                0, -.00783669874902,
    -.01946551099054,  -.0482099128044,  -.0482099128044,  .23336630265161,
                   0,                0,                0,                0,
                   0,                0,                0,                0,
     .00475026273828, -.00783669874902,                0,  .03939732644417,
      .0328943776068,   .0382554606876,   .0382554606876, -.07382466315002,
       .012305588195, -.01946551099054,                0,   .0328943776068,
     .04263002329212,  .05226051095238,  .05226051095238, -.14512177326509,
     .03167368550108,  -.0482099128044,                0,   .0382554606876,
     .05226051095238,  .09096662148872,  .09096662148872, -.32873181469848,
     .03167368550108,  -.0482099128044,                0,   .0382554606876,
     .05226051095238,  .09096662148872,  .09096662148872, -.32873181469848,
    -.16210959536359,  .23336630265161,                0, -.07382466315002,
    -.14512177326509, -.32873181469848, -.32873181469848,  1.5719033807586
    ]).reshape(8,8)

cov_colnames = 'lnpyears smokes 1b.agecat 2.agecat 3.agecat 4.agecat 5.agecat _cons'.split()

cov_rownames = 'lnpyears smokes 1b.agecat 2.agecat 3.agecat 4.agecat 5.agecat _cons'.split()


results_noexposure_constraint2 = Bunch(
                params_table=params_table,
                params_table_colnames=params_table_colnames,
                params_table_rownames=params_table_rownames,
                cov=cov,
                cov_colnames=cov_colnames,
                cov_rownames=cov_rownames,
                **est
                )

est = dict(
           rank = 5,
           N = 10,
           ic = 3,
           k = 7,
           k_eq = 1,
           k_dv = 1,
           converged = 1,
           rc = 0,
           k_autoCns = 1,
           ll = -34.5367006700131,
           k_eq_model = 1,
           df_m = 4,
           chi2 = 554.4168921897579,
           p = 1.1331093797e-118,
           r2_p = np.nan,
           cmdline = "poisson deaths smokes i.agecat, exposure(pyears) constraints(2)",
           cmd = "poisson",
           predict = "poisso_p",
           estat_cmd = "poisson_estat",
           offset = "ln(pyears)",
           gof = "poiss_g",
           chi2type = "Wald",
           opt = "moptimize",
           vce = "oim",
           title = "Poisson regression",
           user = "poiss_lf",
           crittype = "log likelihood",
           ml_method = "e2",
           singularHmethod = "m-marquardt",
           technique = "nr",
           which = "max",
           depvar = "deaths",
           properties = "b V",
          )

params_table = np.array([
     .35978347114582,  .10730668667519,  3.3528522992687,  .00079983377167,
     .14946622996212,  .57010071232952, np.nan,  1.9599639845401,
                   0,                0, np.nan, np.nan,
    np.nan, np.nan, np.nan, np.nan,
     1.9599639845401,                0,  1.4837272702102,  .19510269288329,
     7.6048528509946,  2.852282383e-14,  1.1013330188722,  1.8661215215483,
    np.nan,  1.9599639845401,                0,  2.6270956495127,
     .18372567328363,  14.299012231442,  2.219372691e-46,  2.2669999468414,
      2.987191352184, np.nan,  1.9599639845401,                0,
     3.2898291023835,  .17982035319735,  18.295087535352,  9.055555257e-75,
     2.9373876864294,  3.6422705183376, np.nan,  1.9599639845401,
                   0,  3.7898291023835,  .17982035319735,  21.075640409983,
     1.330935038e-98,  3.4373876864294,  4.1422705183376, np.nan,
     1.9599639845401,                0, -7.9235211042587,  .19177810950798,
    -41.316087245761,                0, -8.2993992919175, -7.5476429165999,
    np.nan,  1.9599639845401,                0]).reshape(7,9)

params_table_colnames = 'b se z pvalue ll ul df crit eform'.split()

params_table_rownames = 'smokes 1b.agecat 2.agecat 3.agecat 4.agecat 5.agecat _cons'.split()

cov = np.array([
     .01151472500521,                0, -.00061274288972, -.00089685568608,
    -.00069335681347, -.00069335681347, -.00921031399899,                0,
                   0,                0,                0,                0,
                   0,                0, -.00061274288972,                0,
     .03806506077031,  .02945948985187,  .02944866089267,  .02944866089267,
    -.02892164840477, -.00089685568608,                0,  .02945948985187,
     .03375512302352,  .02946576868665,  .02946576868665,  -.0286943943397,
    -.00069335681347,                0,  .02944866089267,  .02946576868665,
     .03233535942402,  .03233535942402, -.02885716752919, -.00069335681347,
                   0,  .02944866089267,  .02946576868665,  .03233535942402,
     .03233535942402, -.02885716752919, -.00921031399899,                0,
    -.02892164840477,  -.0286943943397, -.02885716752919, -.02885716752919,
     .03677884328645]).reshape(7,7)

cov_colnames = 'smokes 1b.agecat 2.agecat 3.agecat 4.agecat 5.agecat _cons'.split()

cov_rownames = 'smokes 1b.agecat 2.agecat 3.agecat 4.agecat 5.agecat _cons'.split()


results_exposure_constraint2 = Bunch(
                params_table=params_table,
                params_table_colnames=params_table_colnames,
                params_table_rownames=params_table_rownames,
                cov=cov,
                cov_colnames=cov_colnames,
                cov_rownames=cov_rownames,
                **est
                )


est = dict(
           rank = 5,
           N = 10,
           ic = 3,
           k = 7,
           k_eq = 1,
           k_dv = 1,
           converged = 1,
           rc = 0,
           k_autoCns = 1,
           ll = -34.5367006700131,
           k_eq_model = 1,
           df_m = 4,
           chi2 = 582.5215805315736,
           p = 9.3932644024e-125,
           r2_p = np.nan,
           cmdline = "poisson deaths smokes i.agecat, exposure(pyears) constraints(2) vce(robust)",
           cmd = "poisson",
           predict = "poisso_p",
           estat_cmd = "poisson_estat",
           offset = "ln(pyears)",
           gof = "poiss_g",
           chi2type = "Wald",
           opt = "moptimize",
           vcetype = "Robust",
           vce = "robust",
           title = "Poisson regression",
           user = "poiss_lf",
           crittype = "log pseudolikelihood",
           ml_method = "e2",
           singularHmethod = "m-marquardt",
           technique = "nr",
           which = "max",
           depvar = "deaths",
           properties = "b V",
          )

params_table = np.array([
     .35978347114582,   .1172393358046,  3.0687948603312,  .00214924117257,
      .1299985953974,  .58956834689424, np.nan,  1.9599639845401,
                   0,                0, np.nan, np.nan,
    np.nan, np.nan, np.nan, np.nan,
     1.9599639845401,                0,  1.4837272702102,  .21969092615175,
     6.7537030145039,  1.441186055e-11,  1.0531409672225,  1.9143135731979,
    np.nan,  1.9599639845401,                0,  2.6270956495127,
     .20894895542061,  12.572906355164,  2.975796525e-36,   2.217563222281,
     3.0366280767443, np.nan,  1.9599639845401,                0,
     3.2898291023835,   .2211846822073,  14.873675109654,  4.885611722e-50,
     2.8563150913252,  3.7233431134417, np.nan,  1.9599639845401,
                   0,  3.7898291023835,   .2211846822073,  17.134229479922,
     8.243780087e-66,  3.3563150913252,  4.2233431134417, np.nan,
     1.9599639845401,                0, -7.9235211042587,   .2479876721169,
    -31.951270144281,  5.18748229e-224, -8.4095680102177, -7.4374741982996,
    np.nan,  1.9599639845401,                0]).reshape(7,9)

params_table_colnames = 'b se z pvalue ll ul df crit eform'.split()

params_table_rownames = 'smokes 1b.agecat 2.agecat 3.agecat 4.agecat 5.agecat _cons'.split()

cov = np.array([
      .0137450618599,                0,  .00249770233028,  .00412347653263,
     .00486142402447,  .00486142402447, -.01620342093134,                0,
                   0,                0,                0,                0,
                   0,                0,  .00249770233028,                0,
     .04826410303341,  .04389964215014,  .04391744129373,  .04391744129373,
    -.04609122424924,  .00412347653263,                0,  .04389964215014,
     .04365966597136,  .04367917402468,  .04367917402468, -.04726310745444,
     .00486142402447,                0,  .04391744129373,  .04367917402468,
     .04892266364314,  .04892266364314, -.04794543190806,  .00486142402447,
                   0,  .04391744129373,  .04367917402468,  .04892266364314,
     .04892266364314, -.04794543190806, -.01620342093134,                0,
    -.04609122424924, -.04726310745444, -.04794543190806, -.04794543190806,
     .06149788552196]).reshape(7,7)

cov_colnames = 'smokes 1b.agecat 2.agecat 3.agecat 4.agecat 5.agecat _cons'.split()

cov_rownames = 'smokes 1b.agecat 2.agecat 3.agecat 4.agecat 5.agecat _cons'.split()


results_exposure_constraint2_robust = Bunch(
                params_table=params_table,
                params_table_colnames=params_table_colnames,
                params_table_rownames=params_table_rownames,
                cov=cov,
                cov_colnames=cov_colnames,
                cov_rownames=cov_rownames,
                **est
                )