Dear Statalisters,
given the following dataset ,can anyone help me with what goes on in
minv( ) ,that is to say how I could get " matrix list XZi" ,by hand
especially for more than two independent variables...
Y X Z
1. 1 2 4
2. 2 3 6
3. 3 5 7
4. 4 5 9
. do "C:\WINDOWS\TEMP\STD010000.tmp"
. matrix accum YXZ = Y X Z
(obs=4)
. local nobs = _result(1)
. local nvar = rowsof(YXZ)
. local nvarm1 = `nvar' - 1
. matrix XZ = YXZ[2...,2...]
. matrix Xy = YXZ[1,2...]
. matrix XZi = syminv(XZ)
. matrix b = Xy * XZi
. matrix pv = b * Xy'
. local df = `nobs' - `nvarm1'
. scalar rv = (YXZ[1,1] - pv[1,1])/`df'
. matrix V = XZi * rv
. matrix post b V , dof(`df') obs(`nobs') depname(Y)
. matrix mlout
----------------------------------------------------------------------------
--
Y | Coef. Std. Err. t P>|t| [95% Conf.
Interval]
---------+------------------------------------------------------------------
--
X | .2258065 .1163081 1.941 0.303 -1.252028
1.703641
Z | .4677419 .0838089 5.581 0.113 -.5971512
1.532635
_cons | -1.387097 .2446089 -5.671 0.111 -4.495147
1.720954
----------------------------------------------------------------------------
--
.
end of do-file
. matrix list YXZ
symmetric YXZ[4,4]
Y X Z _cons
Y 30
X 43 63
Z 73 106 182
_cons 10 15 26 4
. matrix list XZ
symmetric XZ[3,3]
X Z _cons
X 63
Z 106 182
_cons 15 26 4
. matrix list XZi
symmetric XZi[3,3]
X Z _cons
X .83870968
Z -.5483871 .43548387
_cons .41935484 -.77419355 3.7096774
.
Thank you very much,
Victor M.Zammit
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
