rout_outlier_test: ROUT Outlier Testing

In OptimModel: Perform Nonlinear Regression Using 'optim' as the Optimization Engine

ROUT Outlier Testing

Description

Perform ROUT outlier testing on rout.fitter object.

Usage

rout_outlier_test(fit, Q = 0.01)

Arguments

fit	A ‘rout.fitter’ object from the rout_fitter function.
Q	Test size for ROUT outlier detection.

Details

rout_outlier_test() is typically called from rout_fitter(), but may also be called directly by the user.

Value

outlier = logical vector. TRUE indicates observation is an outlier via hypothesis testing with unadjust p-values.

outlier.adj = logical vector. TRUE indicates observation is an outlier via hypothesis testing with FDR-adjust p-values.

attr(object, "Q") = test size for outlier detection

Author(s)

Steven Novick

See Also

rout_fitter

Examples

set.seed(123L)

x = rep( c(0, 2^(-4:4)), each=4 )
theta = c(0, 100, log(.5), 2)
y = hill_model(theta, x) + rnorm( length(x), sd=2 )

ii = sample( 1:length(y), 2 )
y[ii] = hill_model(theta, x[ii]) + 5.5*2 + rnorm( length(ii), sd=2 )

fit = rout_fitter(c( theta, log(2) ), hill_model, x=x, y=y, Q=0.01)
rout_outlier_test(fit, Q=0.001)

OptimModel documentation built on May 29, 2024, 9:47 a.m.