Hi, Graham, Thank you for sharing this.
I encountered the same problem. My choice set was unbalanced because some
alternatives were not available for some observations. If I forced the
choice sets balanced, model estimation results would be biased. However,
after including chid.var = "" and alt.va
Hi,
Try this:
Mult3$mode.ids
#[1] 1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6 1 2 3 4 5 6 1
#Levels: 1 2 3 4 5 6
I guess the key is to have equal number of alternatives.
Mult3$mode.ids<- as.numeric(as.character(Mult3$mode.ids))
res<-mlogit.data(Mult3[-25,],shape="long",choice="CHOICE",alt.var="mode.id
-
From: "Leask, Graham"
To: arun
Cc:
Sent: Sunday, April 7, 2013 11:28 AM
Subject: RE: [R] mlogit error
Hi Arun,
I think you are right the issue may be how to identify those obs where this
doesn't apply for whatever reason.
The issue I think is the size of the data which make
On 2011-02-28 12:54, gmacfarlane wrote:
I actually have exactly the same question as Maha, and I wish that Z had
answered the question instead of providing a lesson in streamlined code
(valuable, but not helpful).
I think that Achim answered as best one could, given that Maha
neglected, as you
I actually have exactly the same question as Maha, and I wish that Z had
answered the question instead of providing a lesson in streamlined code
(valuable, but not helpful).
It seems as though mlogit.data repopulates the row.names field, and that
this is where the duplication comes in.
hbwtrips<-
On Fri, 13 Aug 2010, Maha Bakoben wrote:
Hi,
I'm trying to fit a multinomial logistic regression to my data which
consists of 5 discrete variables (scales 1:10) and 1000 observations.
I get the following error:
Error in `row.names<-.data.frame`(`*tmp*`, value = c("NA.NA", "NA.NA", :
duplica
6 matches
Mail list logo
