Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: RE: Breusch and Pagan Lagrangian multiplier test for random effects


From   Caliph Omar Moumin <[email protected]>
To   "[email protected]" <[email protected]>
Subject   Re: st: RE: Breusch and Pagan Lagrangian multiplier test for random effects
Date   Sat, 5 May 2012 08:45:34 -0700 (PDT)

Dear Eric
 
Thank you for your quick reply. I need more???
 
It is not years if you talking about the time period.
In the data "id" is individual patients' identifier.
                 "T" stands how many times each patient is admitted for hospital treatment.
So if i have 10 restrictions; what additional restrictions shall i consider? I don't even know if i have 10 restrictions or not.
are there any important restrictions which i have to consider in panel data with such structure? Could you please explain to me further what you mean?

Kind Regards,
Caliph Omar Moumin

Email:  [email protected] 



----- Original Message -----
From: DE SOUZA Eric <[email protected]>
To: "[email protected]" <[email protected]>
Cc: 
Sent: Saturday, May 5, 2012 5:04 PM
Subject: st: RE: Breusch and Pagan Lagrangian multiplier test for random effects

You introduce 18 new parameters through the A and B matrices. You need to specify at least 12 restrictions on them in order to identify them. You only have 10.
The "years", if they are years and not numbers are only produced when the program breaks. I tried it and got other numbers which also resembled years 
A separate issue: no where in your svar instruction do you make use of the constraints you define


Eric de Souza
College of Europe
Brugge (Bruges), Belgium
http://www.coleurope.eu


-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Caliph Omar Moumin
Sent: 05 May 2012 16:22
To: [email protected]
Subject: st: Breusch and Pagan Lagrangian multiplier test for random effects

Dear all
 
For the past two weeks i spent to decide whether i apply fixed effect or random effect model in my strongly unbalanced panel data. But I couldn't decide  it.
These are the tests i applied so could you please give a minute and advice me what to apply? I understood the my hausman test impllies that i can apply either fixed or random effect modells. Is that so? If that is correct then i choose to apply the random effect model becuase of some time in-variant involved. 
 
What about Breusch-Pagan Lagrange multiplier (LM) test? I have no clue as to how interperate this test? Could any help me?
 
xtdescribe
      id:  6, 9, ..., 809378                                 n =      14503
nadmission1:  1, 2, ..., 16                                  T =         16
           Delta(nadmission1) = 1 unit
           Span(nadmission1)  = 16 periods
           (id*nadmission1 uniquely identifies each observation) Distribution of T_i:   min      5%     25%       50%       75%     95%     max
                         1       1       1         1         1       2      16
     Freq.  Percent    Cum. |  Pattern
 ---------------------------+------------------
    13302     91.72   91.72 |  1...............
      797      5.50   97.21 |  11..............
      160      1.10   98.32 |  111.............
       97      0.67   98.99 |  1111............
       58      0.40   99.39 |  11111...........
       31      0.21   99.60 |  111111..........
       29      0.20   99.80 |  1111111.........
       12      0.08   99.88 |  11111111........
        8      0.06   99.94 |  111111111.......
        9      0.06  100.00 | (other patterns)
 ---------------------------+------------------
    14503    100.00         |  XXXXXXXXXXXXXXXX
 
I want to compare between this two groups xttab group;
                  Overall             Between            Within
    group |    Freq.  Percent      Freq.  Percent        Percent
----------+-----------------------------------------------------
  alcohol |     275      1.64       191      1.32         100.00
 nonalcoh |   16443     98.36     14312     98.68         100.00
----------+-----------------------------------------------------
    Total |   16718    100.00     14503    100.00         100.00
                             (n = 14503)

 
 
.quietly xtreg cost duration sex age group, fe; . estimates store fixed; . quietly xtreg cost duration sex age group, re; . estimates store random; hausman fixed random;
                 ---- Coefficients ----
             |      (b)          (B)            (b-B)     sqrt(diag(V_b-V_B))
             |     fixed        random       Difference          S.E.
-------------+----------------------------------------------------------
-------------+------
    duration |    874.4642     944.5754       -70.11117        84.24204
------------------------------------------------------------------------------
                           b = consistent under Ho and Ha; obtained from xtreg
            B = inconsistent under Ha, efficient under Ho; obtained from xtreg
    Test:  Ho:  difference in coefficients not systematic
                  chi2(1) = (b-B)'[(V_b-V_B)^(-1)](b-B)
                          =        0.69
                Prob>chi2 =      0.4053

 
 Breusch-Pagan Lagrange multiplier (LM)test is performed as follows xtreg cost duration, re; xttest0; Breusch and Pagan Lagrangian multiplier test for random effects
        cost[id,t] = Xb + u[id] + e[id,t]
        Estimated results:
                         |       Var     sd = sqrt(Var)
                ---------+-----------------------------
                    cost |   2.27e+09       47647.13
                       e |   6.78e+08       26038.66
                       u |   1.66e+09       40752.23
        Test:   Var(u) = 0
                              chi2(1) =    59.40
                          Prob > chi2 =     0.0000

A test for heteroskedasticity is performed which shows presence xtreg  cost duration, fe
xttest3

Modified Wald test for groupwise heteroskedasticity in fixed effect regression model
H0: sigma(i)^2 = sigma^2 for all i
chi2 (14503)  = 2.1e+36
Prob>chi2 =      0.0000






Kind Regards,
Moumin

Email:  [email protected] 


*
*  For searches and help try:
*  http://www.stata.com/help.cgi?searchhttp://www.stata.com/support/statalist/faqhttp://www.ats.ucla.edu/stat/stata/

*
*  For searches and help try:
*  http://www.stata.com/help.cgi?searchhttp://www.stata.com/support/statalist/faqhttp://www.ats.ucla.edu/stat/stata/


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index