[Limdep Nlogit List] Maddala's Model-Simultaneous Equations
Laurence Lester
Laurence.Lester at flinders.edu.au
Wed Feb 22 23:16:06 EST 2006
I am trying to set up the code to run Maddala's Model (Simultaneous Equations
with y1 observed and y2 binary). I have taken it from E17.6.2 (and used data
from Limdep (with an added binary constructed) to test the code.
It partly runs, but gives me two mesages. If anyone has the time to run the
attached limdep file they will see the errors. The first appears to be about an
IF statement but there isn't one in the code. The second relates to the final
matrix but I just can't see what the problem is.
If anyone can help I'd appreciate it. The data and code are included below if
that helps.
Regards
Laurence
NILS, Flinders University, Adelaide, SA
read ; nobs = 25 ; nvar = 9 ; names =
COST, SK, SL, SE, SM, PK, PL, PE, PM $
182.373 0.05107 0.24727 0.04253 0.65913 1.00000 1.00000 1.00000 1.00000
183.161 0.05817 0.27716 0.05127 0.61340 1.00270 1.15457 1.30258 1.05525
186.533 0.04602 0.25911 0.05075 0.64411 0.74371 1.15584 1.19663 1.06625
221.710 0.04991 0.24794 0.04606 0.65609 0.92497 1.23535 1.12442 1.12430
255.945 0.05039 0.25487 0.04482 0.64992 1.04877 1.33784 1.25179 1.21694
264.699 0.04916 0.26655 0.04460 0.63969 0.99744 1.37949 1.27919 1.19961
291.160 0.04728 0.26832 0.04369 0.64071 1.00653 1.43458 1.27505 1.19044
274.457 0.05635 0.27167 0.04787 0.62411 1.08757 1.45362 1.30356 1.20612
308.908 0.05258 0.26465 0.04517 0.63760 1.10315 1.51120 1.34277 1.23835
328.286 0.04604 0.26880 0.04576 0.63940 0.99606 1.58186 1.37154 1.29336
338.633 0.05033 0.27184 0.04820 0.62962 1.06321 1.64641 1.38010 1.30703
323.318 0.06015 0.27283 0.04836 0.61886 1.15619 1.67389 1.39338 1.32699
358.435 0.06185 0.27303 0.04563 0.61948 1.30758 1.73430 1.36756 1.30774
366.251 0.05788 0.27738 0.04585 0.61889 1.25413 1.78280 1.38025 1.33946
366.162 0.05903 0.27839 0.04640 0.61617 1.26328 1.81977 1.37630 1.34319
390.668 0.05578 0.28280 0.04530 0.61613 1.26525 1.88531 1.37689 1.34745
412.188 0.05601 0.27968 0.04470 0.61962 1.32294 1.93379 1.34737 1.33143
433.768 0.05452 0.28343 0.04392 0.61814 1.32798 2.00998 1.38969 1.35197
474.969 0.05467 0.27996 0.04114 0.62423 1.40659 2.05539 1.38635 1.37542
521.291 0.05460 0.28363 0.04014 0.62163 1.45100 2.13441 1.40102 1.41878
540.941 0.05443 0.28646 0.04074 0.61837 1.38617 2.20616 1.39197 1.42428
585.447 0.05758 0.28883 0.03971 0.61388 1.49901 2.33869 1.43388 1.43481
630.450 0.05410 0.29031 0.03963 0.61597 1.44957 2.46412 1.46481 1.53356
623.466 0.05255 0.29755 0.04348 0.60642 1.32464 2.60532 1.45907 1.54758
658.235 0.04675 0.28905 0.04479 0.61940 1.20177 2.76025 1.64689 1.54978 $
CREATE ; IF (pl> 1.5) GAMBLE =1; (Else) GAMBLE =0 $ ? Make a Zero/one var
sample; all $
/* Simultanious Equation Models E17-28 Model3 -using limdep data adapted*/
NAMELIST; x1=PE, SK, SL $ ? Explantory for Binary
NAMELIST; x2=PK,PL, SM $ ? Explantory for continuous
CREATE; y1=COST $ ? Continious dependent
CREATE; y2=GAMBLE$ ? Binary choice
PROCECURE = Model3(y1,y2,x1,x2) $
NAMELIST ; xm3 = OR(x1,x2) $
REGRESS ; Lhs=y1; Rhs = xm3; Keep=p1x $
CALC ; t1sq=ssqrd $
PROBIT ; Lhs=y2; Rhs=xm3 $
CREATE ; p2x = xm3'b $
NAMELIST ; z1 = p2x,x1 $
NAMELIST ; z2 = p1x,x2 $
MATRIX ; v0i =<Varb> $
REGRESS ; Lhs=y1; Rhs=z1 $
Matrix ; alpha1=b $
CALC ; gamma1=b(1)
; sigmasq1=ssqrd
; sigma1=s $
PROBIT ; Lhs=y2; Rhs=z2 $
Matrix ; alpha2=b $
CALC ; gamma2=b(1)
; rho12=(2*gamma1*Sqr(sigmasq1))*((1-gamma1*gamma2)^2*t1sq - gamma1^2 -
sigmasq1)
; c=sigmasq1-2*gamma1*rho12*sigma1
; d=gamma2^2 * sigmasq1-2*gamma2*rho12*sigma1 $
Matrix; z1z1i=<z1'z1>
; va1=c*z1z1i + (gamma1^2) * z1z1i * z1'xm3 * <v0i> * xm3'z1 * z1z1i
; v0i=<v0i>
. z2xvi=z2'xm3 * v0i* xm3'z2
; z2xvi=<z2xvi>
; va2=z2xvi+d*z2xvi*z2'xm3*v0i*<xm3'xm3>*v0i*xm3'z2*z2xvi
; Stat(alpha1, va1)
; Stat(alpha2, va2) $
ENDPROCEDURE
EXECUTE; proc=Model3(y1,y2,x1,x2) $
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Simul Model.lim
Type: application/octet-stream
Size: 3288 bytes
Desc: not available
Url : http://limdep.itls.usyd.edu.au/pipermail/limdep/attachments/20060222/906769e9/attachment.obj
More information about the Limdep
mailing list