I'm using plm package to run a twoways within panel regression with a unbalanced data set. I already read about this error but didn't figure out how to solve this problem:
Error in solve.default(vcov(x)[names(coefs_wo_int), names(coefs_wo_int)], :
system is computationally singular: reciprocal condition number = 4.98421e-17
I use 5 independent variables. 3 of them are very large numbers, the other 2 are ratios. Previously, I used the log of the 3 independent variables (x2, x4 and x5)and it worked perfectly without any error messages. Is it possible that it might be due to Multicollinearity and should I actually use the logarithm data set?
fixednormal <- plm(ROA ~ x1 + x2 + x3 + x4 + x5, data=dataset4, model="within", effect="twoways")
fixef(fixednormal, effect="time")
summary(fixednormal)
Changing the unit as a workaround solved the problem.