com.itemanalysis.psychometrics.irt.estimation

Class MstepParallel

java.lang.Object

java.util.concurrent.ForkJoinTask<java.lang.Void>

java.util.concurrent.RecursiveAction

com.itemanalysis.psychometrics.irt.estimation.MstepParallel

All Implemented Interfaces:

java.io.Serializable, java.util.concurrent.Future<java.lang.Void>

public class MstepParallel
extends java.util.concurrent.RecursiveAction

Mstep of the EM algorithm for estimating item parameters in Item Response Theory. Computations can be done in parallel but performance gains appear to be minimal.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
MstepParallel(ItemResponseModel[] irm, DistributionApproximation latentDistribution, EstepEstimates estepEstimates, DensityEstimationType densityEstimationType, int start, int length)

Method Summary

Modifier and Type	Method and Description
protected void	compute() Parallel processing handled here.
protected void	computeDirectly() Mstep computation when the number of items is less than the threshold or when the stopping condition has been reached.
DistributionApproximation	empiricalHistogramLatentDensityEstimation() Nonparametric estimation (empirical histogram method) of the latent density.
int[]	getCodeCount()

Methods inherited from class java.util.concurrent.RecursiveAction

exec, getRawResult, setRawResult

Methods inherited from class java.util.concurrent.ForkJoinTask

adapt, adapt, adapt, cancel, compareAndSetForkJoinTaskTag, complete, completeExceptionally, fork, get, get, getException, getForkJoinTaskTag, getPool, getQueuedTaskCount, getSurplusQueuedTaskCount, helpQuiesce, inForkJoinPool, invoke, invokeAll, invokeAll, invokeAll, isCancelled, isCompletedAbnormally, isCompletedNormally, isDone, join, peekNextLocalTask, pollNextLocalTask, pollTask, quietlyComplete, quietlyInvoke, quietlyJoin, reinitialize, setForkJoinTaskTag, tryUnfork

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

MstepParallel

public MstepParallel(ItemResponseModel[] irm,
                     DistributionApproximation latentDistribution,
                     EstepEstimates estepEstimates,
                     DensityEstimationType densityEstimationType,
                     int start,
                     int length)

Method Detail

computeDirectly

protected void computeDirectly()

Mstep computation when the number of items is less than the threshold or when the stopping condition has been reached. For each item, it uses the optimizer to obtain the estimates that maximize the marginal likelihood.

compute

protected void compute()

Parallel processing handled here.

Specified by:

compute in class java.util.concurrent.RecursiveAction

empiricalHistogramLatentDensityEstimation

public DistributionApproximation empiricalHistogramLatentDensityEstimation()

Nonparametric estimation (empirical histogram method) of the latent density.

Returns:

getCodeCount

public int[] getCodeCount()