r - How to combine a column from a data table to panel data structure -

i have been working large panel data set. simple data set below.

  +--------+------+------+-----+-----+--------+ |  time  | firm | land | lab | cap |   gl   | +--------+------+------+-----+-----+--------+ | jan-00 |    |   25 | 261 |  13 | 161521 | | feb-00 |    |   25 | 334 |  15 | 142452 | | mar-00 |    |   25 | 156 |  14 | 365697 | | apr-00 |    |   28 | 134 |  12 | 355789 | | may-00 |    |   28 | 159 |  15 | 376843 | | jun-00 |    |   28 | 119 |  12 | 258762 | | jul-00 |    |   28 |  41 |  45 | 255447 | | aug-00 |    |   28 | 247 |  75 | 188545 | | sep-00 |    |   28 | 251 |  41 | 213663 | | oct-00 |    |   30 |  62 |  12 | 273209 | | nov-00 |    |   30 | 525 |  15 | 317468 | | dec-00 |    |   30 | 217 |  16 | 238668 | | jan-01 | b    |   42 | 298 |  42 | 241286 | | feb-01 | b    |   42 | 109 |  45 | 135288 | | mar-01 | b    |   42 |   7 |  24 | 363609 | | apr-01 | b    |   42 |  12 |  56 | 318472 | | may-01 | b    |   42 |   0 |  12 | 446279 | | jun-01 | b    |   45 |  50 |  12 | 390230 | | jul-01 | b    |   45 | 143 |  45 | 118945 | | aug-01 | b    |   45 |  85 |  25 | 174887 | | sep-01 | b    |   45 |  80 |  15 | 183770 | | oct-01 | b    |   45 | 214 |  12 | 197832 | | nov-01 | b    |   45 | 525 |  15 | 317468 | | dec-01 | b    |   45 | 217 |  16 | 238668 | +--------+------+------+-----+-----+--------+

the above dataframe can accessed using following codes.

structure(list(time = structure(c(9l, 7l, 15l, 1l, 17l, 13l,  11l, 3l, 23l, 21l, 19l, 5l, 10l, 8l, 16l, 2l, 18l, 14l, 12l,  4l, 24l, 22l, 20l, 6l), .label = c("apr-00", "apr-01", "aug-00",  "aug-01", "dec-00", "dec-01", "feb-00", "feb-01", "jan-00", "jan-01",  "jul-00", "jul-01", "jun-00", "jun-01", "mar-00", "mar-01", "may-00",  "may-01", "nov-00", "nov-01", "oct-00", "oct-01", "sep-00", "sep-01" ), class = "factor"), firm = structure(c(1l, 1l, 1l, 1l, 1l,  1l, 1l, 1l, 1l, 1l, 1l, 1l, 2l, 2l, 2l, 2l, 2l, 2l, 2l, 2l, 2l,  2l, 2l, 2l), .label = c("a", "b"), class = "factor"), land = c(25l,  25l, 25l, 28l, 28l, 28l, 28l, 28l, 28l, 30l, 30l, 30l, 42l, 42l,  42l, 42l, 42l, 45l, 45l, 45l, 45l, 45l, 45l, 45l), lab = c(261l,  334l, 156l, 134l, 159l, 119l, 41l, 247l, 251l, 62l, 525l, 217l,  298l, 109l, 7l, 12l, 0l, 50l, 143l, 85l, 80l, 214l, 525l, 217l ), cap = c(13l, 15l, 14l, 12l, 15l, 12l, 45l, 75l, 41l, 12l,  15l, 16l, 42l, 45l, 24l, 56l, 12l, 12l, 45l, 25l, 15l, 12l, 15l,  16l), gl = c(161521l, 142452l, 365697l, 355789l, 376843l, 258762l,  255447l, 188545l, 213663l, 273209l, 317468l, 238668l, 241286l,  135288l, 363609l, 318472l, 446279l, 390230l, 118945l, 174887l,  183770l, 197832l, 317468l, 238668l)), .names = c("time", "firm",  "land", "lab", "cap", "gl"), class = "data.frame", row.names = c(na,  -24l)) 

i converted data panel structure below.

library(zoo) library(plm)  sys.setlocale("lc_time", "english") dat["time1"] <- as.yearmon(dat$time,format="%b-%y") pdat <-pdata.frame(dat,index=c("firm","time1")) 

original analysis

pdat$cap.lag.ln<-lag(log(pdat$cap), 1) pdat$cap.2lag.ln<-lag(log(pdat$cap), 2) pdat$lab.ln<-log(pdat$lab+1) pdat$lab.lag.ln<-lag(log(pdat$lab+1), 1)  model1<- plm(log(land) ~ cap.lag.ln + cap.2lag.ln + lab.ln + lab.lag.ln,          model = "within", data=pdat) summary(model1)  fv_log <- data.table(model1$model[[1]] - model1$residuals) 

just omit nas in pdat resulted lags combine fitted values

pdat2<-na.omit(pdat) 

the problem

i want combine these fitted values pdat2 need retain panel structure is. if go following cbind panel structure disappeared.

pdat3 <-cbind(pdat2,fv_log) 

even tried omitting old variables , following way.

pdat$cap<-null pdat$lab<-null pdat2 <- na.omit(pdat) pdat2$fv_log <- (model1$model[[1]] - model1$residuals)  , run model2 using data=pdat2 following error message comes up. model2<-plm(log(gl)~ cap.lag.ln + cap.2lag.ln + lab.ln + lab.lag.ln + fv_log, data=pdat2, model = "within")  **error in model.matrix.pformula(formula, data, rhs = 1, model = model,  :  na in individual index variable in addition: warning message: in `[.data.frame`(index, as.numeric(rownames(mf)), ) : nas introduced coercion** 

can please assist me identify problem , run model 2 in plm structure. thank you

updating answer after op provided code needed replication.

i think using data.table create fv_log complicating factor.

i ran following codes after creation of pdat3 above.

# convert fv_log data.frame fv_log<-data.frame(fv_log)  # bind fv_log pdat2 using use pdata.frame preserve structure. pdat3_fix<-pdata.frame(cbind(pdat2,fv_log),index=c("firm","time1"))  

here pdat3_fix, identical pdat2 exception of new column:

note resulting data.frame object of class pdata.frame:

> class(pdat3_fix) [1] "pdata.frame" "data.frame"