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]
st: Variable has replicate levels in nested logit
From
Raluca Prelipceanu <[email protected]>
To
[email protected]
Subject
st: Variable has replicate levels in nested logit
Date
Tue, 4 Dec 2012 15:18:17 +0100
I run the following nlogit and get the error: variable has replicate
levels at the end. Could you help me with this matter?
Thank you.
expand 4
(1347111 observations created)
. sort familyid
. gen alt=mod(_n, 4)
. gen nomig=(alt==1)
. gen sotmig=(alt==2)
. gen sotiemig=(alt==3)
. gen bothmig=(alt==4)
. ta coupmig
coupmig | Freq. Percent Cum.
------------+-----------------------------------
0 | 1,774,208 98.78 98.78
1 | 15,260 0.85 99.63
2 | 2,376 0.13 99.76
3 | 4,304 0.24 100.00
------------+-----------------------------------
Total | 1,796,148 100.00
. gen coupmig2=1 if coupmig==0
(21940 missing values generated)
. replace coupmig2=2 if coupmig==1
(15260 real changes made)
. replace coupmig2=3 if coupmig==2
(2376 real changes made)
. replace coupmig2=4 if coupmig==3
(4304 real changes made)
. gen choice=(coupmig2==alt)
. replace alt=4 if alt==0
(449037 real changes made)
. ta alt
alt | Freq. Percent Cum.
------------+-----------------------------------
1 | 449,037 25.00 25.00
2 | 449,037 25.00 50.00
3 | 449,037 25.00 75.00
4 | 449,037 25.00 100.00
------------+-----------------------------------
Total | 1,796,148 100.00
. gen choice=(coupmig2==alt)
. ta choice
choice | Freq. Percent Cum.
------------+-----------------------------------
0 | 1,347,147 75.00 75.00
1 | 449,001 25.00 100.00
------------+-----------------------------------
Total | 1,796,148 100.00
. nlogitgen typedeg=coupmig2(nomig:1, somemig:2|3|4)
new variable typedeg is generated with 2 groups
label list lb_typedeg
lb_typedeg:
1 nomig
2 somemig
. nlogittree coupmig2 type, choice(choice)
tree structure specified for the nested logit model
typedeg N coupmig2 N k
---------------------------------------------
nomig 1774208 --- 1 1774208 443538
somemig 21940 --- 2 15260 3802
|- 3 2376 591
+- 4 4304 1070
---------------------------------------------
total 1796148 449001
k = number of times alternative is chosen
N = number of observations at each level
nlogit choice sot_age sot_education hychild hschild || type: rural
windex ||coupmig2: nwomenduty acrate, noconstant
> case( ro02a_dwnum)
note: 24218 cases (197748 obs) dropped due to no positive outcome or
multiple positive outcomes per case
variable coupmig2 has replicate levels
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/