ordinalRR.sim {ordinalRR} | R Documentation |
Simulate an ordinal R&R data set.
Description
This function accepts design parameters and model parameters for the random-effects model from Section 3 of Culp, Ryan, Chen, and Hamada (2018) and outputs a simulated R&R data set in preprocessed form.
Usage
ordinalRR.sim(H=4L,I=30L,J=3L,K=2L,mu.a=2.6,sigma.a=.2,lambda=c(11,44,29,40),seed=10L)
Arguments
H |
‘positive integer’ length of the H-point ordinal scale on {1,...,H}. |
I |
‘positive integer’ number of parts. |
J |
‘positive integer’ number of raters. |
K |
‘positive integer’ number of repetitions per rater. |
mu.a |
‘scalar’ mean of log(alpha_j). |
sigma.a |
‘positive scalars’ standard deviation of log(alpha_j). |
lambda |
‘vector of length H of positive scalars’ Dirichlet parameters used to induce a distribution on cutpoints. |
seed |
‘positive integer’ used to set R's seed for pseudo-random number generation. |
Value
I |
‘positive integer’ number of parts. |
J |
‘positive integer’ number of raters. |
K |
‘positive integer’ number of repetitions per rater. |
H |
‘positive integer’ length of the H-point ordinal scale on {1,...,H}. |
x |
‘data.frame’ containing J*K columns and entries from the H-point ordinal scale 1:H. Each part is a row, and there are blocks of K adjacent columns for the repetitions of each of J raters, e.g., rater 1's columns are the first K and rater J's columns are the last K. |
R |
‘array’ x is expanded into a 3-dimensional array (i.e., part 1:I, operator 1:J, ordinal value 1:H) with multinomial counts. |
preprocess |
‘Boolean’ will be TRUE if the data are ready for input into function ordinaRR() for Bayesian analysis with JAGS. |
Author(s)
Ken Ryan
References
Culp, S.L., Ryan, K.J., Chen, J., and Hamada, M.S. (2018). “Analysis of Repeatability and Reproducibility Studies with Ordinal Measurements.” Technometrics, doi:10.1080/00401706.2018.1429317.
See Also
Examples
ordinalRR.sim()