help constraint dialog: manage
-------------------------------------------------------------------------------
Title
[R] constraint -- Define and list constraints
Syntax
Define constraints
constraint [define] # [exp=exp | coeflist]
List constraints
constraint dir [numlist|_all]
constraint list [numlist|_all]
Drop constraints
constraint drop {numlist|_all}
Programmer's commands
constraint get #
constraint free
where coeflist is as defined in [R] test and # is restricted to the range
1 to 1,999, inclusive.
Menu
Statistics > Other > Manage constraints
Description
constraint defines, lists, and drops linear constraints. Constraints are
for use by models that allow constrained estimation.
Constraints are defined by the constraint command. The currently defined
constraints can be listed by either constraint list or constraint dir;
both do the same thing. Existing constraints can be eliminated by
constraint drop.
constraint get and constraint free are programmer's commands. constraint
get returns the contents of the specified constraint in macro r(contents)
and returns in scalar r(defined) 0 or 1 -- 1 being returned if the
constraint was defined. constraint free returns the number of a free
(unused) constraint in macro r(free).
Examples
For use with cnsreg (where there is one equation):
. sysuse auto
. constraint 1 price = weight
. cnsreg mpg price weight, constraints(1)
For use with mlogit (where there are multiple equations):
. webuse sysdsn1
. constraint 2 [Uninsure]2.site = 0
. mlogit insure age male i.site, constraints(2)
. constraint 10 [Indemnity]: 2.site 3.site
. constraint 11 [Indemnity=Prepaid]: 3.site
. mlogit insure age male i.site, constraints(10/11) baseoutcome(3)
In all cases:
. constraint drop 1, 10-11
. constraint drop _all
Also see
Manual: [R] constraint
Help: [R] cnsreg
