agent dunham
2011-Mar-08 10:19 UTC
[R] lm - interaction factors - same number of items per level in factors?
Dear community, Apologies for my questions, I?m pretty newbie. The thing is, I was trying this lm: mod.lm <- lm(df$res ~ df$var1+ df$var2 + df$var3 +fact1:fact2) And factors are like this:> table(fact1)a b c 3 90 83> table(fact2)a b c 4 87 85 Because of the interaction, i obtain the following error, (I guess...):> summary(mod.lm)Residuals: Min 1Q Median 3Q Max -3.0163 -1.1604 -0.1741 0.9281 3.8957 Coefficients: (3 not defined because of singularities) Estimate Std. Error t value Pr(>|t|) (Intercept) 6.5282000 0.7912553 8.250 4.64e-14 *** df$var1 0.0005589 0.0003627 1.541 0.1252 df$var2 -0.2057033 0.0314158 -6.548 6.99e-10 *** df$var3 0.0565537 0.0273348 2.069 0.0401 * fact1a:fact2a NA NA NA NA fact1b:fact2a -0.0086776 1.1375463 -0.008 0.9939 fact15c:fact2a 0.7215114 1.6043880 0.450 0.6535 fact1a:fact2b -1.3091987 0.8251966 -1.587 0.1145 fact1b:fact2b 0.2389941 0.3690889 0.648 0.5182 fact1c:fact2b 0.5016906 0.3474390 1.444 0.1506 fact1a:fact2c NA NA NA NA fact1b:fact2c 0.3121082 0.3491911 0.894 0.3727 fact1c:fact2c NA NA NA NA --- Signif. codes: 0 ?***? 0.001 ?**? 0.01 ?*? 0.05 ?.? 0.1 ? ? 1 Residual standard error: 1.562 on 166 degrees of freedom Multiple R-squared: 0.3526, Adjusted R-squared: 0.3175 F-statistic: 10.04 on 9 and 166 DF, p-value: 2.952e-12 Should a, b, and c in the factors have the same number of entries? Thanks, user at host.com -- View this message in context: http://r.789695.n4.nabble.com/lm-interaction-factors-same-number-of-items-per-level-in-factors-tp3341016p3341016.html Sent from the R help mailing list archive at Nabble.com.
Ista Zahn
2011-Mar-08 17:10 UTC
[R] lm - interaction factors - same number of items per level in factors?
Hi, You should avoid entering interaction terms without also including the lower-order terms involved in the interaction. Also, you are mixing variables from the workspace with those in (the data.frame?) df, which is a very bad idea. Try something like mod.lm <- lm(res ~ var1+ var2 + var3 + fact1*fact2, data=dat) Best, Ista On Tue, Mar 8, 2011 at 5:19 AM, agent dunham <crosspide at hotmail.com> wrote:> Dear community, > > Apologies for my questions, I?m pretty newbie. > > The thing is, I was trying this lm: > > mod.lm <- lm(df$res ~ df$var1+ df$var2 + df$var3 +fact1:fact2) > > And factors are like this: > >> table(fact1) > ?a ?b ?c > ?3 90 83 >> table(fact2) > ?a ?b ?c > ?4 87 85 > > Because of the interaction, i obtain the following error, (I guess...): > >> summary(mod.lm) > > Residuals: > ? ?Min ? ? ?1Q ?Median ? ? ?3Q ? ? Max > -3.0163 -1.1604 -0.1741 ?0.9281 ?3.8957 > > Coefficients: (3 not defined because of singularities) > ? ? ? ? ? ? ? ? ? ? ? ? Estimate Std. Error t value Pr(>|t|) > (Intercept) ? ? ? ? ? ? 6.5282000 ?0.7912553 ? 8.250 4.64e-14 *** > df$var1 ?0.0005589 ?0.0003627 ? 1.541 ? 0.1252 > df$var2 ? -0.2057033 ?0.0314158 ?-6.548 6.99e-10 *** > df$var3 ? ? ? 0.0565537 ?0.0273348 ? 2.069 ? 0.0401 * > fact1a:fact2a ? ? ? ? ?NA ? ? ? ? NA ? ? ?NA ? ? ? NA > fact1b:fact2a ?-0.0086776 ?1.1375463 ?-0.008 ? 0.9939 > fact15c:fact2a ? 0.7215114 ?1.6043880 ? 0.450 ? 0.6535 > fact1a:fact2b ?-1.3091987 ?0.8251966 ?-1.587 ? 0.1145 > fact1b:fact2b ? 0.2389941 ?0.3690889 ? 0.648 ? 0.5182 > fact1c:fact2b ? 0.5016906 ?0.3474390 ? 1.444 ? 0.1506 > fact1a:fact2c ? ? ? ? ?NA ? ? ? ? NA ? ? ?NA ? ? ? NA > fact1b:fact2c ? 0.3121082 ?0.3491911 ? 0.894 ? 0.3727 > fact1c:fact2c ? ? ? ? ?NA ? ? ? ? NA ? ? ?NA ? ? ? NA > --- > Signif. codes: ?0 ?***? 0.001 ?**? 0.01 ?*? 0.05 ?.? 0.1 ? ? 1 > > Residual standard error: 1.562 on 166 degrees of freedom > Multiple R-squared: 0.3526, ? ? Adjusted R-squared: 0.3175 > F-statistic: 10.04 on 9 and 166 DF, ?p-value: 2.952e-12 > > Should a, b, and c in the factors have the same number of entries? > > Thanks, user at host.com > > > -- > View this message in context: http://r.789695.n4.nabble.com/lm-interaction-factors-same-number-of-items-per-level-in-factors-tp3341016p3341016.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. >-- Ista Zahn Graduate student University of Rochester Department of Clinical and Social Psychology http://yourpsyche.org