[R] plm(): observations not used for modelling
David Winsemius
dwinsemius at comcast.net
Tue Nov 6 19:59:11 CET 2012
On Nov 6, 2012, at 7:55 AM, Daniel Bab. wrote:
> Hello,
>
> I have posted this problem before, but thought I try to explain it a bit
> better.
> I'm using the function plm to create a fixed effects model for panel data,
> my method is therefor "within" my effect is "twoways".
> My Data contains unbalanced Panels due to missing Values, but contains 309
> observation for 11 variables (incl. response), with no missing Values. These
> 309 observations distribute over 25 individuals and 18 years.
> The fuction plm only uses 286 of these observations (also if the model is
> changed to first differences and the effect to individual) and omits 23
> observations due to na.action, but in my dataset they do not contain NAs.
> Is this due to the Transformation used in the plm function?
>
> If it is of any help, these are the observations that are omitted (which
> obviously don't contain NAs):
>
> TIME GEO
> 98 1993 Deutschland (einschließlich der ehemaligen DDR seit 1991)
> 99 1994 Deutschland (einschließlich der ehemaligen DDR seit 1991)
> 100 1995 Deutschland (einschließlich der ehemaligen DDR seit 1991)
> 101 1996 Deutschland (einschließlich der ehemaligen DDR seit 1991)
> 370 1999 Österreich
> 372 2001 Österreich
> 375 2004 Österreich
> 385 1995 Polen
> 386 1996 Polen
> 387 1997 Polen
> 495 1991 Schweden
> 439 1992 Slowenien
> 440 1993 Slowenien
> 441 1994 Slowenien
> 442 1995 Slowenien
> 443 1996 Slowenien
> 172 1991 Spanien
> 173 1992 Spanien
> 174 1993 Spanien
> 175 1994 Spanien
> 176 1995 Spanien
> 177 1996 Spanien
> 178 1997 Spanien
> Verkehrstote_Quote Autobahnlaenge_Quote PKW_Quote$Value LKW$Value
I was surprised to a "$" in a variable name. Are you sure that is not the source of your problems?
> dat <- data.frame(a$b = 1:3, d=1:3)
Error: unexpected '=' in "dat <- data.frame(a$b ="
I would not expect the internal parsing routines to necessarily properly handle invalid column names.
--
David.
> 98 123 0.0310283316 479 19.62343
> 99 121 0.0312047562 489 25.99028
> 100 116 0.0313363746 496 27.16505
> 101 107 0.0314931965 501 27.78134
> 370 135 0.0194823002 502 39.96261
> 372 119 0.0196134540 521 41.26695
> 375 108 0.0199949923 505 40.89616
> 385 179 0.0007867343 195 33.66977
> 386 165 0.0008251115 209 35.50949
> 387 189 0.0008443002 221 36.80187
> 495 87 0.0021512832 421 31.19678
> 439 247 0.0125413519 304 16.00871
> 440 247 0.0132326075 317 17.05044
> 441 254 0.0136769861 330 18.09584
> 442 209 0.0144669925 351 20.10579
> 443 195 0.0153063744 366 21.10271
> 172 227 0.0103736290 322 110.86938
> 173 200 0.0128525994 336 67.96822
> 174 163 0.0130329241 343 69.89171
> 175 143 0.0128743969 350 72.00581
> 176 146 0.0137958367 361 74.65096
> 177 139 0.0144557065 374 77.52797
> 178 142 0.0153573304 387 81.11232
> Motorraeder$Value Bevoelkerungsquote Quote_Jung Quote_Alt
> 98 19.561682 226.76063 12.31254 3.928754
> 99 23.297817 227.77846 11.82362 4.011330
> 100 27.802782 228.33996 11.40332 4.087582
> 101 30.189141 229.12098 11.19176 4.026278
> 370 32.947233 95.17546 11.98196 3.378319
> 372 36.778704 95.63432 11.90409 3.558421
> 375 38.808372 97.08449 12.20926 4.066773
> 385 24.079461 123.38487 15.48095 2.154967
> 386 22.688776 123.47698 15.82873 2.102519
> 387 21.791262 123.57274 16.09662 2.027340
> 495 5.238265 19.09182 13.55292 4.302106
> 439 5.502994 98.69708 14.60535 2.364486
> 440 5.516317 98.45870 14.58509 2.464340
> 441 4.523959 98.22782 14.64697 2.535880
> 442 4.523802 98.23123 14.74327 2.612043
> 443 4.019563 98.27018 14.93248 2.567195
> 172 30.199689 77.03350 16.89450 2.962978
> 173 32.074025 77.28903 16.86363 3.062267
> 174 32.684277 77.54355 16.80052 3.163572
> 175 32.817935 77.77117 16.69243 3.261328
> 176 33.068060 77.96193 16.53155 3.352908
> 177 33.171926 78.13598 16.31481 3.438006
> 178 33.548015 78.32325 16.02780 3.510471
> Quote_Erstzulassungen BIP$Value Alkohol.Wert
> 98 0.09782568 21100 13.50
> 99 0.08070474 22200 13.37
> 100 0.08202309 23600 13.35
> 101 0.08530314 23400 13.12
> 370 0.07834963 24900 13.40
> 372 0.07018843 26600 12.80
> 375 0.07575858 28700 12.50
> 385 0.05996661 2800 8.14
> 386 0.07789285 3200 8.08
> 387 0.08464139 3600 8.64
> 495 0.05298563 24200 6.28
> 439 0.05147030 4800 13.64
> 440 0.09212000 5400 14.31
> 441 0.07068413 6100 13.38
> 442 0.08677918 8100 13.36
> 443 0.07988964 8400 12.04
> 172 0.07290907 11400 13.23
> 173 0.07695772 11800 12.50
> 174 0.05522833 10800 12.03
> 175 0.06836836 10800 11.70
> 176 0.06125084 11600 11.38
> 177 0.06563393 12400 11.07
> 178 0.07133359 12800 11.95
>
> TIME and GEO are the Index of my Paneldata, "Verkehrstoten_Quote" is the
> dependent, all others the independent variables.
> If anyone could help me understand, why these observations (or generally
> any) are left out, I would be very glad.
>
> Thank you for dealing with my Problem.
> Regards,
> Daniel
>
>
>
> --
> View this message in context: http://r.789695.n4.nabble.com/plm-observations-not-used-for-modelling-tp4648571.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
David Winsemius, MD
Alameda, CA, USA
More information about the R-help
mailing list