I tried to plot a clustered linear regression model with the cplot command in R
(code below). Leaflet is a binary variable (I know logit would be better),
partisan is nummeric variable (0-4) and partisan_mis a dummy (0,1). As you can
see it is clustered around two variables: around individuals and around the
specific survey. When I try to run the cplot command I always get this error
message:
error in plot.window(...) : need finite 'ylim' values
Zusätzlich: Warnmeldungen: 1: In min(x) : no non-missing arguments to
min; returning Inf
2: In max(x) : no non-missing
arguments to max; returning -Inf
Does anyone know how to solve this problem?
As I suppose that the problem has something to do with the way I coded the
variables, here is how i did it:
voxit$leaflet[voxit$a65== "Ja"] <- 1
voxit$leaflet[voxit$a65== "Nein"] <- 0
voxitf <- subset(voxit, leaflet >= 0 )
voxitf$partisan[voxitf$a87x== "1 Tag oder weniger vorher"] <- 0
voxitf$partisan[voxitf$a87x== "Einige Tage vorher"] <- 1
voxitf$partisan[voxitf$a87x== "1-2 Woche(n) vorher"] <- 2
voxitf$partisan[voxitf$a87x== "Mehrere Wochen vorher"] <- 3
voxitf$partisan[voxitf$a87x== "Schon immer klar"] <- 4
mean(voxitf$partisan, na.rm = TRUE)
voxitf$partisan[is.na(voxitf$a87x)]<- 2.73493
table(voxitf$partisan)
voxitf$partisan_mis[is.na(voxitf$a87x)]<- 1
voxitf$partisan_mis[!is.na(voxitf$a87x)]<- 0
model1.1 <- lm(leaflet ~ partisan + partisan_mis , data = voxitf)
vcov_clust1.1 <- cluster.vcov(model1.1, cbind(voxitf$id,voxitf$projetx))
(marginal_effects1.1 <- margins(model1.1))
summary(marginal_effects1.1)
cplot(model1.1, x = "partisan", dx = "leaflet", what = "effect", se.type =
"shade")
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
