Dear Statalist community,
I am a STATA/MP 14 users trying to deal with the “characteristic contents too long. The maximum value of the contents is 67,784” problem.
More precisely I am running some regression discontinuity and, given that I have to account for crucial fixed effects, I manually add those among the controls. Since factor variables seem not allowed, I must rely on the xi prefix.
I am aware that such xi prefix is obsolete and most of the time brings to troubles, but my guess is that resorting to it is the only solution.
Here a simplified code summarising what just said:
As an extrema ratio, I actually may drop some observations, but I would do that in some further robustness exercise (motivated by other arguments other than computational issues).
Any thought on this issue?
Thank you in advance,
Pasquale
I am a STATA/MP 14 users trying to deal with the “characteristic contents too long. The maximum value of the contents is 67,784” problem.
More precisely I am running some regression discontinuity and, given that I have to account for crucial fixed effects, I manually add those among the controls. Since factor variables seem not allowed, I must rely on the xi prefix.
I am aware that such xi prefix is obsolete and most of the time brings to troubles, but my guess is that resorting to it is the only solution.
Here a simplified code summarising what just said:
Code:
xi:rdrobust Y Running_Variable, c(0) covs(X1 X2 Xn i.year*i.region_code i.id_towns)
Any thought on this issue?
Thank you in advance,
Pasquale
