I have my data-frame Alist500HB1, 10 variables i.e date, cls, log ROA, Log TA etc. I am running a simple panel regression.
plm(Alist500HB1$`Log Q`~Alist500HB1$`CLASSIFIED_BOARD_SYSTEM`+Alist500HB1$`Log TA`+Alist500HB1$`Log ROA`+Alist500HB1$`Log Capex`+Alist500HB1$`onethirdRD`+Alist500HB1$`RD dummy `+Alist500HB1$`onethirdInsider`+Alist500HB1$`PCT_FLT_SHARES_INSTITUTIONS`+Alist500HB1$`onethirdsalesgrowth`+Alist500HB1$`TOT_DEBT_TO_TOT_ASSET`+Alist500HB1$`Logprofit margin`+Alist500HB1$`onethirdassetgrowth`+Alist500HB1$`Log Age`,
data = `Alist500HB1`, index=c("GICS"), model = "random"
)
When I running the regression I am getting this:
Error in [[<-.data.frame(tmp, time.name, value = c(1L, 2L, 3L, 4L, : replacement has 2473 rows, data has 2475
I understand some posts saying this is due to the N/A
data, however, I don't wish to clean them up (if so i can do data <- na.omit(data))
but this will largely reduce my sample size which I don't want to see.
Can someone please give any suggestion how to deal with this? Should I keep cleaning NA
data or other reasons?
Many thanks