e------------------------------------------------------------------------------
help for ^boxcox2^
-------------------------------------------------------------------------------

Box Cox Regression Models
-------------------------
	
	^boxcox2^ depvar [indepvars] [weight] [^if^ exp] [^in^ range ] 
                [^,^ ^model^(^lhs^only |^rhs^only | ^lam^bda | ^theta^)
                ^notr^ans(^varlist^) ^l^evel(^#^) ^lrtest^ 
                ^from^(^init_specs^) ^noc^onstant ^nolog^ ^nologlr^ ]

^fweight^s, and ^iweight^s are allowed ; see help weights.

^boxcox2^ shares the features of all estimation commands; see help @est@ 

The syntax for predict after ^boxcox2^ is 
        ^predict^ [type] newvarname [^if^ exp] [^in^ range] [ ^yhat^ | ^xbt^
        	| ^res^iduals ]

Description
----------

^boxcox2^ finds the maximum likelihood estimates of the parameter[s] of the
Box-Cox transform, the coefficients on the independent variables and the
standard deviation of the normally distributed errors for a model in which
^depvar^ is regressed on ^indepvars^.  The user has the option of
transforming either the ^depvar^, some of the ^indepvars^, both the ^depvar^
and some of the ^indepvars^ with the same transformation parameter, or both
the ^depvar^ and some of the ^indepvars^ with different transformation
parameters.  Any transformed variable must be strictly positive.

Note: this command estimates a superset of the models accommodated by the
^boxcox^ command of official Stata.  See ^[R] boxcox^ for information on the
official command.

The Box-Cox transform of a variable y is given by
			     L
		 (L)	    y  - 1
		y	=   -------
				L
where y must be strictly positive.

Options
-------

^model^( ^lhs^only | ^rhs^only | ^lam^bda | ^theta^ ) specifies which of 
	the four models to fit.  
    
        ^model(lhsonly)^ applies the Box-Cox transform to ^depvar^ only.
        	model(lhsonly) is the default value.

	^model(rhsonly)^ causes the transform to be applied to the
                ^indepvars^ only.

	^model(lambda)^ causes the transform to be aplied to both
                ^depvar^ and ^indepvars^, and they are transformed by the same
                parameter.

	^model(theta)^ causes the transform to be aplied to both
                ^depvar^ and ^indepvars^, but this time each side is
                tranformed by a separate parameter.


^notrans^(^varlist^) specifies that the variables in   
	^varlist^ are to be included as non-transformed independent
	variables.

^lrtest^ specifies that a likelihood ratio test is to be performed and
	reported for each independent variable.


^from()^ allows the user to specify the initial values for Box-Cox
	transformation parameter(s).  
        
	Model		Initial value specification
	-------------------------------------------
	^lhsonly^	from( #_t, copy) 
	^rhsonly^  	from( #_l, copy) 
	^lambda^  	from( #_l, copy)
	^theta^  	from( #_l #_t, copy)
	-------------------------------------------
	
	where #_l is the intial value of lambda and #_t is the initial value
	of theta.

^noconstant^ suppresses the constant term (intercept) in the model.

^nolog^ suppresses the iteration log when estimating the full model.

^nologlr^ suppresses the iteration log when estimating the 
	restricted models required by the ^lrtest^ option.  If 
	^nologlr^ is specified when ^lrtest^ is not, then it is ignored.

^level^(^#^)} specifies the confidence level, in percent, for
    confidence intervals. The default is ^level(95)^ or as set by 
    ^set^ ^level^; see 23.5 Specifying the width of confidence
    intervals.  

Options for predict
-------------------

^xbt^ the default, calculates the transformed linear prediction. 

^yhat^ calculates the predicted value of y.

^residuals^ calculates the residuals left over after the predicted value
	of  y has been subtracted from the actual value.


Examples
--------

. ^boxcox2 mpg weight price, notrans(foreign) model(theta) lrtest ^
. ^boxcox2 mpg weight price, model(lambda) lrtest^ 
. ^boxcox2 mpg weight price, notrans(foreign) model(rhs)^ 
. ^boxcox2 mpg weight price, notrans(foreign) model(lhsonly) lrtest^

Also see
--------

Manual: ^[U] 23 Estimation and post-estimation commands^,
	^[U] 29 Overview of model estimation in Stata^,
	^[R] boxcox^
	^[R] lnskew0^
On-line: help for ^est^, ^lincom^, ^lnskew0^, ^postest^, ^predict^, ^regress^, 
        ^test^, ^testnl^