Hello, I'm writing my Bachelor Thesis and I'm using the package plm for a panelregression with fixed effects (individual and time). My Dataset contains 11 variables (including the response) and has 486 observations of these. In the observation I have missing values, which don't make a problem. If I reduce my dataset from observations with missing values, it still contains 309 observations. My Problem is, that my model by using plm only uses 286 of my observations. 23 observations are missing, which are distributed over 6 of my individuals and don't contain missing values. None of these observations should be identical to another in the dataset (I only checked for one), so I can't figure out why I'm missing these observations. Example from my Dataset: TIME GEO Verkehrstote_Quote Autobahnlaenge_Quote PKW_Quote$Value 20 1991 Belgien 188 0.05069903 398 21 1992 Belgien 167 0.05094485 401 22 1993 Belgien 165 0.05115993 408 23 1994 Belgien 168 0.05119066 417 24 1995 Belgien 143 0.05119066 422 25 1996 Belgien 134 0.05143647 428 LKW$Value Motorraeder$Value Bevoelkerungsquote Quote_Jung Quote_Alt 20 36.04695 14.91943 306.8666 13.79864 3.537448 21 36.61945 16.16444 307.9428 13.47332 3.595531 22 37.24554 17.38125 309.3661 13.21728 3.680058 23 38.71045 18.61270 310.3589 13.01022 3.738638 24 39.68186 19.74222 311.2790 12.82610 3.812795 25 41.11191 20.90102 311.6622 12.65081 3.808343 Quote_Erstzulassungen BIP$Value Alkohol.Wert 20 0.01177985 16300 11.49 21 0.12079010 17300 11.74 22 0.09550608 18300 11.72 23 0.09595748 19600 11.23 24 0.08662883 21500 11.10 25 0.09409657 21400 11.07 Has anyone seen this problem before? Thank you for dealing with my problem. Regards, Daniel -- View this message in context: http://r.789695.n4.nabble.com/I-m-missing-observations-in-plm-tp4647791.html Sent from the R help mailing list archive at Nabble.com.