estheta: A estimation theta function.

In takuizum/irtfun2: Some useful functions for IRT

Description

A estimation theta function.

Usage

estheta(
  xall,
  param,
  est = "EAP",
  nofrands = 10,
  method = "NR",
  file = "default",
  output = FALSE,
  IDc = 1,
  gc = 2,
  fc = 3,
  gh = TRUE,
  N = 31,
  D = 1.702,
  maxtheta = 4,
  mintheta = -4,
  mu = 0,
  sigma = 1,
  sampling_engine = "rejection_Cpp",
  warm_up = 100
)

Arguments

xall	a item response data
param	a item parameter file.If class is df, parameter column must be "a", "b" and "c". If class is matrix, you should set a parameter in first column, b parameter in second and c in third.
est	estimation method option. "EAP","MAP","MLE","PVs."
nofrands	the number of PVs.
method	iteration method option in MLE and MAP. "NR" is Newton Rapthon, "SANN" is Simulated Aannealing, "Brent" is optimization using optimise function
file	a name of output csv file.
output	logical. If TRUE write csv file. defauli is FALSE.
IDc	colomn of ID, default is 1.Even If df has no ID columns this function works.
gc	column of group or population numbers. If df has no grouo ID, set 0.
fc	column of first item response, default is 3.
gh	logical. If TRUE, is default, gaussHermite quadrature runs for EAP estimation.
N	the number of nodes. default is 31.
D	a factor constant. deafault is 1.702.
maxtheta	the maximul value of theta in integration.
mintheta	the minimum value of theta in integration.
mu	a hyperparameter of prior distribution.
sigma	same as above.
sampling_engine	an option of sampling engine. if select "rejection_R", the rejection samplimg method is conducted and this engine loop witten in R lang this is so somewhat slow, but "rejection_Cpp" is for loop in C++ lang so very fast. If select "slice_R" , the slice sampling method will be conducted.
warm_up	the number of iteration times for warm up in slice sampling

Value

a list has ID, rawscore, theta, se, person fit index Z3 and log likelihood.

Author(s)

Takumi, Shibuya., Daisuke, Ejiri., Tadashi, Shibayama. in Tohoku University.

Examples

set.seed(123)
theta <- rnorm(200)
b <- rnorm(15)
a <- rlnorm(15,0,0.5)
c <- rep(0,15)
data <- irtfun2::sim_gen(theta,a,b,c)
para <- cbind(a,b,c)
pv <- estheta(data, param = para, est = "PVs", gc = 0, fc = 2)

head(pv$OVs_df)

takuizum/irtfun2 documentation built on May 10, 2020, 8:30 a.m.