Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • poisson modelling

    Hi Statalist,

    I used following data to do poisson modelling:

    Code:
    Year M_all_P M_G_High_L1_AUse_RR15_C_adj M_all_L1_AUse_RR15_C_adj
    1987 5520001 211.9177 501.407
    1988 5633569 216.2777 511.7229
    1989 5755082 221.1636 523.2832
    1990 5867299 225.8139 534.2859
    1991 5973605 230.4786 545.3229
    1992 6074767 234.9648 555.9374
    1993 6155299 239.3794 566.3824
    1994 6236709 243.6229 576.4227
    1995 6325682 250.3768 592.4029
    1996 6417271 260.5307 616.4274
    1997 6502261 273.0925 646.1493
    1998 6581291 282.8546 669.2469
    1999 6664289 289.1082 684.043
    2000 6749964 298.3964 706.0193
    2001 6847125 313.2063 741.0602
    2002 6952335 336.035 795.0741
    2003 7057470 350.3287 828.8935
    2004 7158720 364.5615 862.5689
    2005 7269239 382.8875 905.9291
    2006 7391066 400.3706 947.295
    2007 7547610 415.8048 983.8129
    2008 7724059 429.8253 1016.986
    2009 7915795 440.6469 1042.59
    2010 8064259 448.9114 1062.145
    2011 8199360 456.432 1079.939
    2012 8356709 465.1911 1100.663
    2013 5792143 322.43 762.8839
    Code:
    . poisson M_all_L1_AUse_RR15_C_adj Year, exposure(M_all_P)
    note: you are responsible for interpretation of noncount dep. variable
    
    Iteration 0:   log likelihood = -127.10332
    Iteration 1:   log likelihood = -127.10331
    
    Poisson regression                              Number of obs     =         27
                                                    LR chi2(1)        =     431.15
                                                    Prob > chi2       =     0.0000
    Log likelihood = -127.10331                     Pseudo R2         =     0.6291
    
    ------------------------------------------------------------------------------------------
    M_all_L1_AUse_RR15_C_adj |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
    -------------------------+----------------------------------------------------------------
                        Year |   .0191066   .0009262    20.63   0.000     .0172912     .020922
                       _cons |  -47.34115   1.854261   -25.53   0.000    -50.97544   -43.70687
                 ln(M_all_P) |          1  (exposure)
    ------------------------------------------------------------------------------------------
    Code:
    . poisson M_G_High_L1_AUse_RR15_C_adj Year , exposure(M_all_P)
    note: you are responsible for interpretation of noncount dep. variable
    
    Iteration 0:   log likelihood = -107.80178
    Iteration 1:   log likelihood = -107.80178
    
    Poisson regression                              Number of obs     =         27
                                                    LR chi2(1)        =     182.22
                                                    Prob > chi2       =     0.0000
    Log likelihood = -107.80178                     Pseudo R2         =     0.4580
    
    ---------------------------------------------------------------------------------------------
    M_G_High_L1_AUse_RR15_C_adj |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
    ----------------------------+----------------------------------------------------------------
                           Year |   .0191066   .0014247    13.41   0.000     .0163142    .0218991
                          _cons |  -48.20238   2.852216   -16.90   0.000    -53.79262   -42.61214
                    ln(M_all_P) |          1  (exposure)
    ---------------------------------------------------------------------------------------------
    I found that coefficients for Year are exactly the same.

    I'm wondering if there is any explanation for this? why the coefficients are exactly the same?

    Regards,
    Last edited by Masoumeh Sanagou; 05 Nov 2017, 18:26.

  • #2
    I tried your first model on your data and did not get the same results


    Code:
    . poisson m_all_l1_ause_rr15_c_adj year, exp(m_all_p)
    note: you are responsible for interpretation of noncount dep. variable
    
    
    Poisson regression                              Number of obs     =         27
                                                    LR chi2(1)        =       5.14
                                                    Prob > chi2       =     0.0234
    Log likelihood = -1901.0491                     Pseudo R2         =     0.0013
    
    ------------------------------------------------------------------------------------------
    m_all_l1_ause_rr15_c_adj |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
    -------------------------+----------------------------------------------------------------
                        year |   .0022039   .0009729     2.27   0.023     .0002971    .0041107
                       _cons |  -13.64075   1.946652    -7.01   0.000    -17.45611   -9.825379
                 ln(m_all_p) |          1  (exposure)
    ------------------------------------------------------------------------------------------

    Comment


    • #3
      Thanks for the reply.

      I tried one more time and got the previous results.


      Code:
      . poisson m_all_l1_ause_rr15_c_adj year, exposure(m_all_p)
      note: you are responsible for interpretation of noncount dep. variable
      
      Iteration 0:   log likelihood = -127.10331 
      Iteration 1:   log likelihood = -127.10331 
      
      Poisson regression                              Number of obs     =         27
                                                      LR chi2(1)        =     431.15
                                                      Prob > chi2       =     0.0000
      Log likelihood = -127.10331                     Pseudo R2         =     0.6291
      
      ------------------------------------------------------------------------------------------
      m_all_l1_ause_rr15_c_adj |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
      -------------------------+----------------------------------------------------------------
                          year |   .0191066   .0009262    20.63   0.000     .0172912     .020922
                         _cons |  -47.34115   1.854261   -25.53   0.000    -50.97544   -43.70687
                   ln(m_all_p) |          1  (exposure)
      ------------------------------------------------------------------------------------------

      Comment


      • #4
        Originally posted by Masoumeh Sanagou View Post
        why the coefficients are exactly the same?
        .ÿ
        .ÿclearÿ*

        .ÿ
        .ÿquietlyÿinputÿintÿYearÿlongÿM_all_Pÿdouble(M_G_High_L1_AUse_RR15_C_adjÿM_all_L1_AUse_RR15_C_adj)

        .ÿ
        .ÿgenerateÿdoubleÿratioÿ=ÿM_all_L1_AUse_RR15_C_adjÿ/ÿM_G_High_L1_AUse_RR15_C_adj

        .ÿsummarizeÿratio

        ÿÿÿÿVariableÿ|ÿÿÿÿÿÿÿÿObsÿÿÿÿÿÿÿÿMeanÿÿÿÿStd.ÿDev.ÿÿÿÿÿÿÿMinÿÿÿÿÿÿÿÿMax
        -------------+---------------------------------------------------------
        ÿÿÿÿÿÿÿratioÿ|ÿÿÿÿÿÿÿÿÿ27ÿÿÿÿ2.366045ÿÿÿÿ4.17e-07ÿÿÿ2.366044ÿÿÿ2.366046

        .ÿ
        .ÿexit

        endÿofÿdo-file


        .


        Comment

        Working...
        X