Skip to content

Problem with PVCM: "0 (non-NA) cases"

1 message · Johannes Habel

Johannes Habel
# 
Hello everybody,

I have a problem with the the function PVCM in the plm package:

I have an unbalanced panel with over 200 companies and 1 to 14 years for
each company. I already excluded all NAs:
Now I'm trying to fit a variable coefficients model:
This results in the following error message: "Fehler in lm.fit(x, y, offset
= offset, singular.ok = singular.ok, ...) : 0 (non-NA) cases"

I can't figure out what's wrong... Can somebody help me, please?

Thanks so much!

Johannes

PS: Here's additional information on my dataset:
Classes ?plm.dim? and 'data.frame':     1531 obs. of  4 variables:
 $ company: Factor w/ 210 levels "1164","1177",..: 1 1 1 1 1 1 1 1 1 1 ...
 $ year   : Factor w/ 16 levels "1993","1994",..: 2 3 4 5 6 7 8 9 10 11 ...
 $ v1     : num  0.0993 0.1488 0.0604 0.0902 0.0573 ...
 $ v2     : int  75 75 78 72 74 73 70 70 70 67 ...
 - attr(*, "na.action")=Class 'omit'  Named int [1:979] 12 13 14 15 16 17 18
26 39 40 ...
  .. ..- attr(*, "names")= chr [1:979] "12" "13" "14" "15" ...
company          year           v1                 v2       
 3121   :  15   2007   :136   Min.   :-1.05746   Min.   :49.00  
 8762   :  15   2006   :128   1st Qu.: 0.08706   1st Qu.:71.00  
 1382   :  14   2005   :125   Median : 0.12834   Median :76.00  
 1388   :  14   2003   :119   Mean   : 0.13399   Mean   :75.37  
 1440   :  14   2004   :119   3rd Qu.: 0.18909   3rd Qu.:80.00  
 1663   :  14   2002   :115   Max.   : 0.71280   Max.   :90.00  
 (Other):1445   (Other):789