rmsd.yai: Root Mean Square Difference between observed and imputed

In yaImpute: Nearest Neighbor Observation Imputation and Evaluation Tools

Root Mean Square Difference between observed and imputed

Description

Computes the root mean square difference (RMSD) between observed and imputed values for each observation that has both. RMSD is computationally like RMSE, but they differ in interpretation. The RMSD values can be scaled to afford comparisons among variables.

Usage

rmsd.yai (object,vars=NULL,scale=FALSE,...)

Arguments

object	an object created by yai or impute.yai
vars	a list of variable names you want to include, if NULL all available variables are included
scale	when TRUE, the values are scaled (see details), if a named vector, the values are scaled by the corresponding values.
...	passed to called methods, very useful for passing argument ancillaryData to function impute.yai

Details

By default, RMSD is computed using standard formula for its related statistic, RMSE. When scale=TRUE, or set of values is supplied, RMSD is divided by the scaling factor. The scaling factor is the standard deviation of the reference observations under the assumption that they are representative of the population.

Value

A data frame with the row names as vars and the column as rmsd. When scale=TRUE, the column name is rmsdS. The scaling factors used, if any, are returned as an attribute.

Author(s)

Nicholas L. Crookston [e-mail]
Andrew O. Finley [e-mail]

See Also

yai, impute.yai and doi: 10.18637/jss.v023.i10.

yaImpute documentation built on Nov. 4, 2022, 1:06 a.m.