validate.Rq.RdThe validate function when used on an object created by
Rq does resampling validation of a quantile regression
model, with or without backward step-down variable deletion. Uses
resampling to estimate the optimism in various measures of predictive
accuracy which include mean absolute prediction error (MAD), Spearman
rho, the \(g\)-index, and the intercept and slope
of an overall
calibration \(a + b\hat{y}\). The "corrected"
slope can be thought of as shrinkage factor that takes into account
overfitting. validate.Rq can also be used when a model for a
continuous response is going to be applied to a binary response. A
Somers' \(D_{xy}\) for this case is computed for each resample by
dichotomizing y. This can be used to obtain an ordinary receiver
operating characteristic curve area using the formula \(0.5(D_{xy} +
1)\). See predab.resample for information about confidence limits
and for the list of resampling methods.
The LaTeX needspace package must be in effect to use the
latex method.
# fit <- fitting.function(formula=response ~ terms, x=TRUE, y=TRUE)
# S3 method for class 'Rq'
validate(fit, method="boot", B=40,
bw=FALSE, rule="aic", type="residual", sls=0.05, aics=0,
force=NULL, estimates=TRUE, pr=FALSE, u=NULL, rel=">",
tolerance=1e-7, ...)a fit derived by Rq. The options x=TRUE and y=TRUE
must have been specified. See validate for a description of
arguments method - pr.
see
validate and predab.resample and
fastbw
If specifed, y is also dichotomized at the cutoff u for
the purpose of getting a bias-corrected estimate of \(D_{xy}\).
relationship for dichotomizing predicted y. Defaults to
">" to use y>u. rel can also be "<",
">=", and "<=".
ignored
other arguments to pass to predab.resample, such as group, cluster, and subset
matrix with rows corresponding to various indexes, and optionally \(D_{xy}\), and columns for the original index, resample estimates, indexes applied to whole or omitted sample using model derived from resample, average optimism, corrected index, and number of successful resamples.
prints a summary, and optionally statistics for each re-fit
set.seed(1)
x1 <- runif(200)
x2 <- sample(0:3, 200, TRUE)
x3 <- rnorm(200)
distance <- (x1 + x2/3 + rnorm(200))^2
f <- Rq(sqrt(distance) ~ rcs(x1,4) + scored(x2) + x3, x=TRUE, y=TRUE)
#> Warning: Solution may be nonunique
#Validate full model fit (from all observations) but for x1 < .75
validate(f, B=20, subset=x1 < .75) # normally B=300
#> Warning: Solution may be nonunique
#> Warning: Solution may be nonunique
#> Warning: Solution may be nonunique
#> Warning: Solution may be nonunique
#> Warning: Solution may be nonunique
#> Warning: Solution may be nonunique
#> index.orig training test optimism index.corrected n
#> MAD 0.618 0.6109 0.642 -0.0312 0.649 20
#> rho 0.254 0.2760 0.208 0.0677 0.186 20
#> g 0.198 0.2807 0.190 0.0906 0.107 20
#> Intercept 0.155 0.0973 0.295 -0.1979 0.353 20
#> Slope 0.815 0.8827 0.672 0.2107 0.604 20
#Validate stepwise model with typical (not so good) stopping rule
validate(f, B=20, bw=TRUE, rule="p", sls=.1, type="individual")
#>
#> Backwards Step-down - Original Model
#>
#> Deleted Chi-Sq d.f. P Residual d.f. P AIC
#> x3 0.38 1 0.5382 0.38 1 0.5382 -1.62
#> x2 3.24 3 0.3565 3.62 4 0.4605 -4.38
#> x1 5.49 3 0.1391 9.11 7 0.2449 -4.89
#>
#> Approximate Estimates after Deleting Factors
#>
#> Coef S.E. Wald Z P
#> [1,] 0.9961 0.07502 13.28 0
#>
#> Factors in Final Model
#>
#> None
#> Warning: Solution may be nonunique
#> Warning: Solution may be nonunique
#> Warning: Solution may be nonunique
#> Warning: Solution may be nonunique
#> Warning: Solution may be nonunique
#> Warning: 1 non-positive fis
#> Warning: Solution may be nonunique
#> Warning: 1 non-positive fis
#> Warning: Solution may be nonunique
#> Warning: 2 non-positive fis
#> Warning: Solution may be nonunique
#> Warning: 2 non-positive fis
#> Warning: 2 non-positive fis
#> Warning: Solution may be nonunique
#> Warning: Solution may be nonunique
#> Warning: Solution may be nonunique
#> Warning: Solution may be nonunique
#> Warning: Solution may be nonunique
#> Warning: Solution may be nonunique
#> Warning: 1 non-positive fis
#> Warning: Solution may be nonunique
#> Warning: Solution may be nonunique
#> Warning: Solution may be nonunique
#> Warning: Solution may be nonunique
#> Warning: Solution may be nonunique
#> Warning: 1 non-positive fis
#> Warning: Solution may be nonunique
#> Warning: Solution may be nonunique
#> Warning: 4 non-positive fis
#> Warning: Solution may be nonunique
#> Warning: 4 non-positive fis
#> index.orig training test optimism index.corrected n
#> MAD 0.688 0.615 0.666 -0.0513 0.7390 20
#> rho 0.000 0.326 0.259 0.0672 -0.0672 20
#> g 0.000 0.352 0.253 0.0994 -0.0994 20
#> Intercept 0.000 0.000 0.264 -0.2636 0.2636 20
#> Slope 1.000 1.000 0.757 0.2432 0.7568 20
#>
#> Factors Retained in Backwards Elimination
#>
#> x1 x2 x3
#> * *
#> * *
#> * *
#> * *
#> * *
#> *
#> * * *
#> * * *
#> * *
#> *
#> * *
#> * *
#> * * *
#> *
#> * * *
#> *
#> *
#> *
#> *
#> * * *
#>
#> Frequencies of Numbers of Factors Retained
#>
#> 1 2 3
#> 7 8 5