.-
help for ^swaic^                                                  (STB-54: sg134)
.-

Stepwise model selection using AIC
--------------------------------------------------------

	^swaic^ [^,^ ^fp(^format^)^ ^fc^hi^(^format^)^ ^m^odel ^b^ack ]

Description
-----------

 ^swaic^ performs automatic model selection using Akaike information 
 criterion (AIC). AIC=-2*log-likelihood + 2 * (number of parameters)


Options
-------
^fp(^format^)^ specifies the output format for p values, default is %9.4f. 

^fc(^format^)^ specifies the output format for chi2 values, default is %9.2f. 

^model^ reports the model which reachs the minimum AIC with either forward
    or backward method.

^back^ specifies a backward method, the default is a forward method.


Examples
--------

   . xi: logit outcome age sex i.expose hibp bmi
   . swaic, m

   . swaic, b m

   . stset time outcome
   . xi: stcox age sex bmi hibp i.expose, nolog
   . swaic, b m

   . xi:streg age sex bmi hibp i.expose, dist(weibull)
   . swaic, b m

   . xi: poisson outcome age sex bmi hibp i.expose, e(time) nolog
   . swaic, m


Author
------

    Zhiqiang Wang
    Menzies School of Health Research
    Darwin, Australia
    wang@@menzies.edu.au


Also see
--------

 Manual:  ^[U] 23 Estimation and post-estimation commands^
	  ^[R] lrtest^
     
On-line:  help for @linktest@, @test@, @testnl@, @lrtest2@ (if installed) @streg@