24 May
2004
24 May
'04
10:41 a.m.
Dear all,
I'm trying to use Zelig to automatically calculate predicted
probabilities for all individual observation pixels (=10.027), but get
an error message after the sim() statement..
I have 17 significant variables in the model.
Does anybody know what could be wrong?
-------------------------------------------------------------------------------
Any help would be appreciated,
Thanks,
Tom
--
Tom Vanwalleghem
Laboratory for experimental geomorphology
Redingenstraat 16
B-3000 LEUVEN
+32(0)16/326414
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)latte.harvard.edu
[un]subscribe Options: http://lists.hmdc.harvard.edu/?info=zelig
Zelig program information: http://gking.harvard.edu/zelig/
z.out1<-zelig(depressies~archeodist + aspect2 +
aspect3 + B_1 + B_2 +
B_3 + B_4 + B_5 + B_6 + B_7 + G_fb + G_zbr + G_zle + gully +
roadgully
+ helling + wegen, data = data1 , model = "relogit" , tau = 2338/556823
, bias.correct = TRUE)
x.out1<-setx(z.out1,data=data1,cond=TRUE)
s.out1<-sim(z.out1,x=x.out1)
Error in coef %*% t(x) : non-conformable arguments
24 May
24 May
11:34 a.m.
Dear Tom,
Could you please send the traceback()? It's essential for
diagnosing problems.
In addition, does your data frame contain variables not included
in the specification? If so, you might want to subset out just
the variables you're using and see if you get the same error.
Thanks, Olivia.
----- Original Message -----
From: "Tom Vanwalleghem" <Tom.Vanwalleghem(a)geo.kuleuven.ac.be>
To: <zelig(a)latte.harvard.edu>
Sent: Monday, May 24, 2004 10:41 AM
Subject: [zelig] problem with sim
Dear all,
I'm trying to use Zelig to automatically calculate predicted
probabilities for all individual observation pixels (=10.027),
but get
an error message after the sim() statement..
I have 17 significant variables in the model.
Does anybody know what could be wrong?
--------------------------------------------------------------
-----------------
>z.out1<-zelig(depressies~archeodist + aspect2 + aspect3 + B_1
+ B_2 +
B_3 + B_4 + B_5 + B_6 + B_7 + G_fb + G_zbr + G_zle +
gully +
roadgully
+ helling + wegen, data = data1 , model =
"relogit" , tau =
2338/556823
, bias.correct = TRUE)
Any help would be appreciated,
Thanks,
Tom
--
Tom Vanwalleghem
Laboratory for experimental geomorphology
Redingenstraat 16
B-3000 LEUVEN
+32(0)16/326414
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)latte.harvard.edu
[un]subscribe Options:
http://lists.hmdc.harvard.edu/?info=zelig
x.out1<-setx(z.out1,data=data1,cond=TRUE)
s.out1<-sim(z.out1,x=x.out1)
Error in coef %*% t(x) : non-conformable arguments
Zelig program information:
http://gking.harvard.edu/zelig/
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)latte.harvard.edu
[un]subscribe Options: http://lists.hmdc.harvard.edu/?info=zelig
Zelig program information: http://gking.harvard.edu/zelig/
25 May
25 May
3:59 a.m.
Could you please send the traceback()? It's
essential for diagnosing
problems.
traceback()
3: qi.glm(object, simpar, x = x, x1
= x1)
2: sim.relogit(z.out1, x = x.out1)
1: sim(z.out1, x = x.out1)
In addition, does your data frame contain variables
not included in
the specification?
The data frame did contain variables not included in the specification,
but even after deleting these variables (through dataframe$var<-NULL), I
keep getting the same error.
Additionally, X.OUT1 seems to be a regular 10.027*19 matrix
dim(x.out1)
[1] 10027 19
mode(x.out1)
[1] "numeric
Any further suggestions?
Best regards,
Tom
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)latte.harvard.edu
[un]subscribe Options: http://lists.hmdc.harvard.edu/?info=zelig
Zelig program information: http://gking.harvard.edu/zelig/
26 May
26 May
1:02 a.m.
Dear Tom,
Thanks for identifying this bug in the sim() procedure for
relogit. If you download and install 1.0-7, it should run.
Also, you might be able to pare down your synatx a little bit.
x.out <- setx(z.out, cond = TRUE)
is sufficient for in-sample inference. We're revising the
setx() function so that cond = TRUE indicates in-sample
inference. You should leave out the data argument, if that's
what you're trying to do.
Yours,
Olivia.
----- Original Message -----
From: "Tom Vanwalleghem" <Tom.Vanwalleghem(a)geo.kuleuven.ac.be>
To: "Olivia Lau" <olau(a)fas.harvard.edu>
Cc: <zelig(a)latte.harvard.edu>du>; "Kosuke Imai"
<kimai(a)Princeton.Edu>
Sent: Tuesday, May 25, 2004 3:59 AM
Subject: Re: [zelig] problem with sim
included in
specification,
> Could you please send the traceback()? It's
essential for
diagnosing
problems.
traceback()
3: qi.glm(object, simpar, x =
x, x1 = x1)
2: sim.relogit(z.out1, x = x.out1)
1: sim(z.out1, x = x.out1)
> In addition, does your data frame contain variables not the
specification?
The data frame did contain variables not included in the but even after deleting these variables (through
dataframe$var<-NULL), I
keep getting the same error.
Additionally, X.OUT1 seems to be a regular 10.027*19 matrix
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)latte.harvard.edu
[un]subscribe Options: http://lists.hmdc.harvard.edu/?info=zelig
Zelig program information: http://gking.harvard.edu/zelig/
dim(x.out1)
[1] 10027 19
mode(x.out1)
[1] "numeric
Any further suggestions?
Best regards,
Tom
24 May
24 May
4:12 p.m.
Could you also let us know whether x.out1 is a proper matrix?
Thanks,
Kosuke
---------------------------------------------------------
Kosuke Imai Office: Corwin Hall 041
Assistant Professor Phone: 609-258-6601 (Direct)
Department of Politics eFax: 973-556-1929 (Direct)
Princeton University Email: kimai(a)Princeton.Edu
Princeton, NJ 08544-1012 http://www.princeton.edu/~kimai
---------------------------------------------------------
On Mon, 24 May 2004, Tom Vanwalleghem wrote:
Dear all,
I'm trying to use Zelig to automatically calculate predicted
probabilities for all individual observation pixels (=10.027), but get
an error message after the sim() statement..
I have 17 significant variables in the model.
Does anybody know what could be wrong?
-------------------------------------------------------------------------------
Any help would be appreciated,
Thanks,
Tom
--
Tom Vanwalleghem
Laboratory for experimental geomorphology
Redingenstraat 16
B-3000 LEUVEN
+32(0)16/326414
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)latte.harvard.edu
[un]subscribe Options: http://lists.hmdc.harvard.edu/?info=zelig
Zelig program information: http://gking.harvard.edu/zelig/
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)latte.harvard.edu
[un]subscribe Options: http://lists.hmdc.harvard.edu/?info=zelig
Zelig program information: http://gking.harvard.edu/zelig/
z.out1<-zelig(depressies~archeodist + aspect2 +
aspect3 + B_1 + B_2 +
B_3 + B_4 + B_5 + B_6 + B_7 + G_fb + G_zbr + G_zle + gully +
roadgully
+ helling + wegen, data = data1 , model = "relogit" , tau = 2338/556823
, bias.correct = TRUE)
x.out1<-setx(z.out1,data=data1,cond=TRUE)
s.out1<-sim(z.out1,x=x.out1)
Error in coef %*% t(x) : non-conformable arguments
7298
days inactive
7300
days old
4 comments
3 participants
participants (3)
-
Kosuke Imai -
Olivia Lau -
Tom Vanwalleghem