Displaying 3 results from an estimated 3 matches for "adjr".
Did you mean:
addr
2005 Sep 12
3
Covert list of list to dataframe for export or outputting by(test) output
...1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2)), .Label =
c("Acenaphthene",
"Cadmium, Total"), class = "factor"), AdjRes = c(0.03, 0.24,
0.0082, 0.29, 0.01, 0.19, 0.2, 0.22, 0.0032, 0.32, 0.09,
0.3, 0.0061, 0.0037, 0.38, 0.2, 0.36, 0.09, 0.77, 0.2, 0.19,
0.2, 0.18, 0.6, 0.32, 0.34, 1.5, 1.2, 0.21, 0.21, 0.01, 0.2,
0.32, 0.2, 0.19, 0.02, 0.26, 0.37, 0.18, 1.3, 0.0088, 4.78,
0.0033, 0.32, 2.1, 0.1...
2005 Jun 17
2
adjusted R^2 vs. ordinary R^2
I thought the point of adjusting the R^2 for degrees of
freedom is to allow comparisons about goodness of fit between
similar models with different numbers of data points. Someone
has suggested to me off-list that this might not be the case.
Is an ADJUSTED R^2 for a four-parameter, five-point model
reliably comparable to the adjusted R^2 of a four-parameter,
100-point model? If such values
2007 Dec 19
0
leaps
...Multiple R-Squared: 0.9932, Adjusted R-squared: 0.9914
F-statistic: 549.8 on 18 and 68 DF, p-value: < 2.2e-16*
newmat <- cosmat[,-c(5,8,10,12,13,15,16,18,20:44)]
newmat <- cbind(newmat,sinmat[,-c(7:11,13:44)])
Regmod <- regsubsets(newmat, yy)
rs <- summary(Regmod)
which.max(rs$adjr)
*[1] 8
*rs$which[which.max(rs$adjr), ]
*(Intercept) cos1 cos2 cos3 cos4 cos6
TRUE TRUE TRUE FALSE TRUE FALSE
cos7 cos9 cos11 cos14 cos17 cos19
TRUE FALSE FALSE...