Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: component order for fmm
From
Scott Merryman <[email protected]>
To
[email protected]
Subject
Re: st: component order for fmm
Date
Tue, 21 Aug 2012 07:24:51 -0500
Why does the order of the components matter? Also, you should mention
that Partha Deb's -fmm- and Joerg Luedicke's -fmmlc- are from ssc.
One way would be to change the initial values. For example:
clear
set seed 891234
set obs 3
gen id = _n
expand 100
gen y = rnormal(50,10) if id == 1
replace y = rnormal(100,10) if id == 2
replace y = rnormal(150,10) if id == 3
fmm y , components(3) mix(normal) from(50 100 150 1 1 1 1 1)
fmm y , components(3) mix(normal) from(50 150 100 1 1 1 1 1)
fmm y , components(3) mix(normal) from(150 100 50 1 1 1 1 1)
Scott
On Mon, Aug 20, 2012 at 7:15 PM, yannan shen <[email protected]> wrote:
> Hi all,
> I am working on some panel data (stocks by month)
> Every year, I am using the fmm package to identify stocks return into
> groups. And then I am using the fmmlc, savec savep command to save up
> the result.
> My code looks like this
>
> keep if obs==67 *obs 67 =06/1998
> fmm ro_cons, components(2) mix(normal)
> fmmlc, savec savep
> fmm ro_cons, components(3) mix(normal)
> fmmlc, savec savep
>
> what I care is _b[_cons] the coefficient for the intercept.
> However, I noticed that for some years _b[_cons group1]>_b[_cons
> group2], and sometimes, it's the other way around.
> It there anyway I can put a constraint to this command and make sure
> the intercept term for group 1 is always greater than intercept for
> group2?
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/