FindBestBandwidth1Iter: One iteration of method to find optimal bandwidth (between...

In CambridgeAssessmentResearch/KernEqWPS: Kernel Equating Without Pre-Smoothing

One iteration of method to find optimal bandwidth (between 0.05 and 30) that minimses cross-validation error between two half-samples.

Description

Based on paper by Liang and von Davier (2014). It works by splitting the data set in half, fitting a continuized distribution with a gaussian kernel to one half of the data, and then evaluating the log-likelihood of this distribution against the other half of the data.

Usage

FindBestBandwidth1Iter(scores, fX)

Arguments

scores	Vector of possible scores.
fX	Number of people achieving each score (should be whole numbers). Must be same length as scores.

Value

The function returns a single value representing the optimal bandwidth.

References

Liang, T., & von Davier, A. A. (2014). Cross-validation: An alternative bandwidth-selection method in kernel equating. Applied Psychological Measurement, 38(4), 281-295.

Examples

tot1=rowSums(mathsdata)
tab1=table(tot1)
FindBestBandwidth1Iter(as.numeric(names(tab1)),as.vector(tab1))

CambridgeAssessmentResearch/KernEqWPS documentation built on Feb. 23, 2024, 9:34 p.m.