Irm3PL

java.lang.Object
- com.itemanalysis.psychometrics.irt.model.AbstractItemResponseModelWithGradient
- - com.itemanalysis.psychometrics.irt.model.Irm3PL

All Implemented Interfaces:

ItemResponseModel

Direct Known Subclasses:

IrmBinary
```
public class Irm3PL
extends AbstractItemResponseModelWithGradient
```
An implementation of AbstractItemResponseModel that allows for the three-parameter logistic (3PL) model, two-parameter logistic (2PL) model, one-parameter logistic (1PL) model, and the Rasch model. The particular type of item response model is determined by the constructor used to create the object. See the constructors for more details on which one to use for each type of model.

Field Summary
- Fields inherited from class com.itemanalysis.psychometrics.irt.model.AbstractItemResponseModelWithGradient
  groupId, isFixed, maxCategory, maxWeight, minCategory, minWeight, name, ncat, scoreWeight

Constructor Summary

Constructors
Constructor and Description
`Irm3PL(double difficulty, double D)` Constructor for one parameter logistic model
`Irm3PL(double discrimination, double difficulty, double D)` Constructor for two parameter logistic model.
`Irm3PL(double discrimination, double difficulty, double guessing, double D)` Constructor for three parameter logistic model

Method Summary

All Methods Instance Methods Concrete Methods
Modifier and Type	Method and Description
`double`	`acceptAllProposalValues()` Proposal values for every item parameter are obtained at each iteration of the estimation routine.
`double`	`addPriorsToLogLikelihood(double loglike, double[] iparam)` Adds prior probabilities to the loglikelihood.
`double[]`	`addPriorsToLogLikelihoodGradient(double[] loglikegrad, double[] iparam)` Adds log-prior probabilities to the item loglikelihood.
`double`	`cumulativeProbability(double theta, int category)` Not implemented.
`double`	`derivTheta(double theta)` Derivative from Mathematica.
`double`	`derivTheta2(double theta, int response)` From Equating recipes.
`double`	`expectedValue(double theta)` Computes the expected value
`double`	`getDifficulty()` Gets the item difficulty parameter.
`ItemParamPrior`	`getDifficultyPrior()`
`double`	`getDifficultyStdError()` Gets the item difficulty standard error.
`double`	`getDiscrimination()` Gets item discrimination.
`ItemParamPrior`	`getDiscriminationPrior()`
`double`	`getDiscriminationStdError()` Gets the standard error for the item discrimination estimate.
`double`	`getGuessing()` Gets the pseudo-guessing (i.e.
`ItemParamPrior`	`getGuessingPrior()`
`double`	`getGuessingStdError()` Gets the guessing parameter estimate standard error.
`double[]`	`getItemParameterArray()`
`int`	`getNumberOfEstimatedParameters()` Number of estimated parameters does not count any values fixed during estimation.
`int`	`getNumberOfParameters()` Gets the number of item parameters.
`double`	`getProposalDifficulty()` A proposal difficulty value is obtained during each iteration of the estimation routine.
`double`	`getProposalDiscrimination()`
`double`	`getScalingConstant()`
`double`	`getSlipping()` Gets the slipping (i.e.
`double`	`getSlippingStdError()` Gets the slipping parameter estimate standard error.
`double[]`	`getStepParameters()` Polytomous item response models may have step parameters.
`double[]`	`getStepStdError()` Gets that standard errors for each step parameter estimate.
`double[]`	`getThresholdParameters()` Polytomous item response models may use an overall item difficulty parameter and two or more threshold parameters.
`double[]`	`getThresholdStdError()` Gets the array of standard errors fort eh threshold parameter estimates.
`IrmType`	`getType()` Gets the type of item response model.
`double[]`	`gradient(double theta, double[] iparam, int k, double D)` Computes the gradientAt (vector of first partial derivatives) with respect to the item parameters.
`double[]`	`gradient(double theta, int k)` Computes gradientAt using item parameters stored in the object.
`double[][]`	`hessian(double theta)` Hessian or matrix of second derivatives.
`void`	`incrementMeanMean(org.apache.commons.math3.stat.descriptive.moment.Mean meanDiscrimination, org.apache.commons.math3.stat.descriptive.moment.Mean meanDifficulty)` Mean/mean linking coefficients are computed from the mean item difficulty and mean item discrimination.
`void`	`incrementMeanSigma(org.apache.commons.math3.stat.descriptive.moment.Mean mean, org.apache.commons.math3.stat.descriptive.moment.StandardDeviation sd)` Mean/sigma linking coefficients are computed from the mean and standard deviation of item difficulty.
`double`	`itemInformationAt(double theta)` Computes the item information function at theta.
`double[]`	`nonZeroPrior(double[] param)` If the prior density for a parameter is zero, adjust parameter to the nearest non zero value.
`double`	`probability(double theta, double[] iparam, int response, double D)` Comute the probability of a correct response.
`double`	`probability(double theta, int response)` Computes the probability of response.
`void`	`scale(double intercept, double slope)` Linear transformation of item parameters.
`void`	`setDifficulty(double difficulty)` Set difficulty parameter to an existing value.
`void`	`setDifficultyPrior(ItemParamPrior difficultyPrior)`
`void`	`setDifficultyStdError(double stdError)` Item difficulty standard error may be computed external to the class.
`void`	`setDiscrimination(double discrimination)` Set discrimination parameter to an existing value.
`void`	`setDiscriminationPrior(ItemParamPrior discriminationPrior)`
`void`	`setDiscriminationStdError(double stdError)` The standard error may be computed external to the class.
`void`	`setGuessing(double guessing)` Set lower asymptote parameter to an existing value.
`void`	`setGuessingPrior(ItemParamPrior guessingPrior)`
`void`	`setGuessingStdError(double stdError)` The guessing parameter standard error may be computed external to the class.
`void`	`setProposalDifficulty(double proposalDifficulty)` A proposal difficulty value is obtained during each iteration of the estimation routine.
`void`	`setProposalDiscrimination(double discrimination)` Set the proposed discrimination estimate.
`void`	`setProposalGuessing(double guessing)` A proposal guessing parameter value is obtained during each iteration of the estimation routine.
`void`	`setProposalSlipping(double slipping)` A proposal slipping parameter value is obtained during each iteration of the estimation routine.
`void`	`setProposalStepParameters(double[] step)`
`void`	`setProposalThresholds(double[] thresholds)` Sets the proposed threshold parameters estimates to particular values.
`void`	`setSlipping(double slipping)` Set upper asymptote parameter to an existing value.
`void`	`setSlippingPrior(ItemParamPrior slippingPrior)`
`void`	`setSlippingStdError(double slipping)` The slipping parameter standard error may be computed external to the class.
`void`	`setStandardErrors(double[] x)`
`void`	`setStepParameters(double[] step)`
`void`	`setStepPriorAt(ItemParamPrior prior, int index)`
`void`	`setStepStdError(double[] stdError)` Sets the standard error for the step parameter estimates.
`void`	`setThresholdParameters(double[] thresholdParameters)` Sets the threshold parameters to particular values.
`void`	`setThresholdStdError(double[] stdError)` Set the threshold standard errors.
`java.lang.String`	`toString()` A string representation of the item parameters.
`double`	`tSharpExpectedValue(double theta, double intercept, double slope)` Computes probability of a response under a linear transformation.
`double`	`tSharpProbability(double theta, int response, double intercept, double slope)` Computes probability of a response under a linear transformation.
`double`	`tStarExpectedValue(double theta, double intercept, double slope)` Computes item expected value under a linear transformation.
`double`	`tStarProbability(double theta, int response, double intercept, double slope)` Computes probability of a response under a linear transformation.

Methods inherited from class com.itemanalysis.psychometrics.irt.model.AbstractItemResponseModelWithGradient
defaultScoreWeights, getGroupId, getItemFitStatistic, getItemScoring, getMaxScoreWeight, getMinScoreWeight, getName, getNcat, getScoreWeights, isFixed, setFixed, setGroupId, setItemFitStatistic, setItemScoring, setName, setScoreWeights

Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

- Constructor Detail
  - Irm3PL
```
public Irm3PL(double discrimination,
              double difficulty,
              double guessing,
              double D)
```
    Constructor for three parameter logistic model
    
    Parameters:
    
    discrimination - item discrimination parameter
    
    difficulty - item difficulty parameter
    
    guessing - lower-asymptote parameter
    
    D - scaling factor
  - Irm3PL
```
public Irm3PL(double discrimination,
              double difficulty,
              double D)
```
    Constructor for two parameter logistic model.
    
    Parameters:
    
    discrimination - item discrimination parameter
    
    difficulty - item difficulty parameter
    
    D - scaling factor
  - Irm3PL
```
public Irm3PL(double difficulty,
              double D)
```
    Constructor for one parameter logistic model
    
    Parameters:
    
    difficulty - item difficulty parameter
    
    D - scaling factor
- Method Detail
  - probability
```
public double probability(double theta,
                          double[] iparam,
                          int response,
                          double D)
```
    Comute the probability of a correct response. This method uses a combination of stored parameters and those passed in the iparam array. This arrangement allows for fixing some item parameters during estimation. For example, instantiating this object as a 2PL and setting the stored value of the guessing parameter will result in a 3PL with a fixed guessing parameter. If you do not want to use the stored values of any parameter, then the iparam array should be of length = 3 and contain the discrimination, difficulty, and guessing parameters, respectively. The order of parameters in iparam is as follows: 1. Rasch model: iparam[0] = difficulty 2. 2PL: iparam[0] = discrimination, iparam[1] = difficulty 3. 3PL: iparam[1] = discrimination, iparam[1] = difficulty, iparam[2] = guessing
    
    Parameters:
    
    theta - person ability parameter.
    
    iparam - item parameter array that is one to three in length.
    
    D - scaling constant.
    
    response - an item response category.
    
    Returns:
    
    probability of a correct answer.
  - probability
```
public double probability(double theta,
                          int response)
```
    Description copied from interface: ItemResponseModel
    
    Computes the probability of response. It ranges between 0 and 1.
    
    Parameters:
    
    theta - a person ability value.
    
    response - an item response (i.e. a person's score on an item).
    
    Returns:
    
    the probability of the response.
  - expectedValue
```
public double expectedValue(double theta)
```
    Computes the expected value
    
    Parameters:
    
    theta - a person ability value.
    
    Returns:
  - cumulativeProbability
```
public double cumulativeProbability(double theta,
                                    int category)
```
    Not implemented. Only implemented for the graded response model.
    
    Parameters:
    
    theta - a person ability value
    
    category - response category
    
    Returns:
  - gradient
```
public double[] gradient(double theta,
                         double[] iparam,
                         int k,
                         double D)
```
    Computes the gradientAt (vector of first partial derivatives) with respect to the item parameters. This method uses item parameters passed to the method. It does NOT use item parameters stored in the object. Note: The second argument (int k) is not actually used by this class. It is here to satisfy the interface.
    
    Parameters:
    
    theta - person ability estimate.
    
    iparam - array of item parameters.
    
    k - response category
    
    D - scaling constant that is either 1 or 1.7
    
    Returns:
    
    an array of first partial derivatives (i.e. the gradientAt).
  - gradient
```
public double[] gradient(double theta,
                         int k)
```
    Computes gradientAt using item parameters stored in the object.
    
    Parameters:
    
    theta - person ability value
    
    k - response category
    
    Returns:
    
    gradientAt
  - hessian
```
public double[][] hessian(double theta)
```
    Hessian or matrix of second derivatives.
    
    Parameters:
    
    theta - person ability value.
    
    Returns:
    
    a two-way array containing the Hessian matrix values.
  - derivTheta2
```
public double derivTheta2(double theta,
                          int response)
```
    From Equating recipes. Computes the derivative with respect to person ability.
    
    Parameters:
    
    theta - person ability value.
    
    response - item response value.
    
    Returns:
    
    derivative wrt theta.
  - derivTheta
```
public double derivTheta(double theta)
```
    Derivative from Mathematica. First derivative with respect to person ability.
    
    Parameters:
    
    theta - person ability value.
    
    Returns:
    
    first derivative wrt theta.
  - itemInformationAt
```
public double itemInformationAt(double theta)
```
    Computes the item information function at theta.
    
    Parameters:
    
    theta - person ability value.
    
    Returns:
    
    item information.
  - nonZeroPrior
```
public double[] nonZeroPrior(double[] param)
```
    Description copied from interface: ItemResponseModel
    
    If the prior density for a parameter is zero, adjust parameter to the nearest non zero value.
  - setDiscriminationPrior
```
public void setDiscriminationPrior(ItemParamPrior discriminationPrior)
```
  - setDifficultyPrior
```
public void setDifficultyPrior(ItemParamPrior difficultyPrior)
```
  - setGuessingPrior
```
public void setGuessingPrior(ItemParamPrior guessingPrior)
```
  - setSlippingPrior
```
public void setSlippingPrior(ItemParamPrior slippingPrior)
```
  - setStepPriorAt
```
public void setStepPriorAt(ItemParamPrior prior,
                           int index)
```
  - getDiscriminationPrior
```
public ItemParamPrior getDiscriminationPrior()
```
  - getDifficultyPrior
```
public ItemParamPrior getDifficultyPrior()
```
  - getGuessingPrior
```
public ItemParamPrior getGuessingPrior()
```
  - addPriorsToLogLikelihood
```
public double addPriorsToLogLikelihood(double loglike,
                                       double[] iparam)
```
    Description copied from interface: ItemResponseModel
    
    Adds prior probabilities to the loglikelihood. This method is used in {@link com.itemanalysis.psychometrics.irt.estimation.MarginalMaximumLikelihoodEstimation
  - addPriorsToLogLikelihoodGradient
```
public double[] addPriorsToLogLikelihoodGradient(double[] loglikegrad,
                                                 double[] iparam)
```
    Description copied from interface: ItemResponseModel
    
    Adds log-prior probabilities to the item loglikelihood. This method is used in MarginalMaximumLikelihoodEstimation.
  - incrementMeanSigma
```
public void incrementMeanSigma(org.apache.commons.math3.stat.descriptive.moment.Mean mean,
                               org.apache.commons.math3.stat.descriptive.moment.StandardDeviation sd)
```
    Mean/sigma linking coefficients are computed from the mean and standard deviation of item difficulty. The summary statistics are computed in a storeless manner. This method allows for the incremental update to item difficulty summary statistics by combining them with other summary statistics.
    
    Parameters:
    
    mean - item difficulty mean.
    
    sd - item difficulty standard deviation.
  - incrementMeanMean
```
public void incrementMeanMean(org.apache.commons.math3.stat.descriptive.moment.Mean meanDiscrimination,
                              org.apache.commons.math3.stat.descriptive.moment.Mean meanDifficulty)
```
    Mean/mean linking coefficients are computed from the mean item difficulty and mean item discrimination. The summary statistics are computed in a storeless manner. This method allows for the incremental update to item difficulty summary statistics by combining them with other summary statistics.
    
    Parameters:
    
    meanDiscrimination - item discrimination mean.
    
    meanDifficulty - item difficulty mean.
  - tStarProbability
```
public double tStarProbability(double theta,
                               int response,
                               double intercept,
                               double slope)
```
    Computes probability of a response under a linear transformation. This method is mainly used for the characteristic curve linking methods (see StockingLordMethod). It applies the linear transformation such that the New form is transformed to the Old Form.
    
    Parameters:
    
    theta - examinee proficiency value
    
    response - target category
    
    intercept - linking coefficient for intercept
    
    slope - linking coefficient for slope
    
    Returns:
  - tSharpProbability
```
public double tSharpProbability(double theta,
                                int response,
                                double intercept,
                                double slope)
```
    Computes probability of a response under a linear transformation. This method is mainly used for the characteristic curve linking methods (see StockingLordMethod). It applies the linear transformation such that the Old form is transformed to the New Form.
    
    Parameters:
    
    theta - examinee proficiency value
    
    response - target category
    
    intercept - linking coefficient for intercept
    
    slope - linking coefficient for slope
    
    Returns:
  - tStarExpectedValue
```
public double tStarExpectedValue(double theta,
                                 double intercept,
                                 double slope)
```
    Computes item expected value under a linear transformation. This method is mainly used for the characteristic curve linking methods (see StockingLordMethod). It applies the linear transformation such that the New form is transformed to the Old Form.
    
    Parameters:
    
    theta - person ability value
    
    intercept - intercept linking coefficient.
    
    slope - slope linking coefficient.
    
    Returns:
    
    expected value under a linear transformation.
  - tSharpExpectedValue
```
public double tSharpExpectedValue(double theta,
                                  double intercept,
                                  double slope)
```
    Computes probability of a response under a linear transformation. This method is mainly used for the characteristic curve linking methods (see StockingLordMethod). It applies the linear transformation such that the Old form is transformed to the New Form.
    
    Parameters:
    
    theta - examinee proficiency value
    
    intercept - linking coefficient for intercept
    
    slope - linking coefficient for slope
    
    Returns:
    
    expected value under a linear transformation.
  - getNumberOfParameters
```
public int getNumberOfParameters()
```
    Gets the number of item parameters.
    
    Returns:
    
    number of item parameters.
  - getNumberOfEstimatedParameters
```
public int getNumberOfEstimatedParameters()
```
    Description copied from interface: ItemResponseModel
    
    Number of estimated parameters does not count any values fixed during estimation.
    
    Returns:
  - scale
```
public void scale(double intercept,
                  double slope)
```
    Linear transformation of item parameters.
    
    Parameters:
    
    intercept - intercept transformation coefficient.
    
    slope - slope transformation coefficient.
  - getItemParameterArray
```
public double[] getItemParameterArray()
```
  - setStandardErrors
```
public void setStandardErrors(double[] x)
```
  - getType
```
public IrmType getType()
```
    Gets the type of item response model.
    
    Returns:
    
    type of item response model.
  - getDifficulty
```
public double getDifficulty()
```
    Gets the item difficulty parameter.
    
    Returns:
    
    item difficulty.
  - setDifficulty
```
public void setDifficulty(double difficulty)
```
    Set difficulty parameter to an existing value. If you are using this method to fix an item parameter during estimation, you must also set the proposal value in setProposalDifficulty(double).
  - getProposalDifficulty
```
public double getProposalDifficulty()
```
    A proposal difficulty value is obtained during each iteration of the estimation routine. This method gets the proposal item difficulty values. This method is needed for estimating item difficulty.
    
    Returns:
    
    proposed item difficulty value.
  - setProposalDifficulty
```
public void setProposalDifficulty(double proposalDifficulty)
```
    A proposal difficulty value is obtained during each iteration of the estimation routine. This method sets the proposal value.
    
    Parameters:
    
    proposalDifficulty - proposed item difficulty value.
  - getDifficultyStdError
```
public double getDifficultyStdError()
```
    Gets the item difficulty standard error.
    
    Returns:
    
    item difficulty standard error.
  - setDifficultyStdError
```
public void setDifficultyStdError(double stdError)
```
    Item difficulty standard error may be computed external to the class. This method sets the difficulty standard error to a computed value.
    
    Parameters:
    
    stdError - item difficulty standard error.
  - getDiscrimination
```
public double getDiscrimination()
```
    Gets item discrimination.
    
    Returns:
    
    item discrimination.
  - setDiscrimination
```
public void setDiscrimination(double discrimination)
```
    Set discrimination parameter to an existing value. If you are using this method to fix an item parameter during estimation, you must also set the proposal value with setProposalDiscrimination(double).
  - getProposalDiscrimination
```
public double getProposalDiscrimination()
```
  - setProposalDiscrimination
```
public void setProposalDiscrimination(double discrimination)
```
    Set the proposed discrimination estimate.
    
    Parameters:
    
    discrimination - proposed item discrimination value.
  - getDiscriminationStdError
```
public double getDiscriminationStdError()
```
    Gets the standard error for the item discrimination estimate.
    
    Returns:
    
    item discrimination standard error.
  - setDiscriminationStdError
```
public void setDiscriminationStdError(double stdError)
```
    The standard error may be computed external to the class. It can be set to a specific value with this method.
    
    Parameters:
    
    stdError - item discrimination standard error.
  - getGuessing
```
public double getGuessing()
```
    Gets the pseudo-guessing (i.e. lower asymptote) parameter.
    
    Returns:
    
    guessing parameter.
  - setGuessing
```
public void setGuessing(double guessing)
```
    Set lower asymptote parameter to an existing value. If you are using this method to fix an item parameter during estimation, you must also set the proposal value in setProposalGuessing(double).
  - setProposalGuessing
```
public void setProposalGuessing(double guessing)
```
    A proposal guessing parameter value is obtained during each iteration of the estimation routine. This method sets the proposal value.
    
    Parameters:
    
    guessing - proposed guessing parameter estimate.
  - getGuessingStdError
```
public double getGuessingStdError()
```
    Gets the guessing parameter estimate standard error.
    
    Returns:
    
    guessing parameter estimate standard error.
  - setGuessingStdError
```
public void setGuessingStdError(double stdError)
```
    The guessing parameter standard error may be computed external to the class. Use this method to set the standard error to a particular value.
    
    Parameters:
    
    stdError - standard error for the guessing parameter estimate.
  - setSlipping
```
public void setSlipping(double slipping)
```
    Description copied from interface: ItemResponseModel
    
    Set upper asymptote parameter to an existing value. If you are using this method to fix an item parameter during estimation, you must also set the proposal value in ItemResponseModel.setProposalSlipping(double).
  - setProposalSlipping
```
public void setProposalSlipping(double slipping)
```
    Description copied from interface: ItemResponseModel
    
    A proposal slipping parameter value is obtained during each iteration of the estimation routine. This method sets the proposal value.
    
    Parameters:
    
    slipping - proposed slipping parameter estimate.
  - setSlippingStdError
```
public void setSlippingStdError(double slipping)
```
    Description copied from interface: ItemResponseModel
    
    The slipping parameter standard error may be computed external to the class. Use this method to set the standard error to a particular value.
    
    Parameters:
    
    slipping - standard error for the slipping parameter estimate.
  - getSlipping
```
public double getSlipping()
```
    Description copied from interface: ItemResponseModel
    
    Gets the slipping (i.e. upper asymptote) parameter.
    
    Returns:
    
    slipping parameter.
  - getSlippingStdError
```
public double getSlippingStdError()
```
    Description copied from interface: ItemResponseModel
    
    Gets the slipping parameter estimate standard error.
    
    Returns:
    
    slipping parameter estimate standard error.
  - getScalingConstant
```
public double getScalingConstant()
```
  - acceptAllProposalValues
```
public double acceptAllProposalValues()
```
    Proposal values for every item parameter are obtained at each iteration of the estimation routine. The proposal values for each parameters are obtained for each in turn using the estimated values from the previous iteration. For example, a proposal difficulty estimate for itemA is obtained in iteration k+1 using estimates from iteration k. Then, a proposal difficulty estimate for itemB is obtained in iteration k+1 using estimates from iteration k (even though a new estimate exists for itemA). After obtaining proposal values for every item on the test, the proposal values can be accepted as the new parameter estimates. This method must be called to accept the proposal values as the new estimates. Returns the maximum relative absolute difference between existing parameters and new parameters.
  - getStepParameters
```
public double[] getStepParameters()
```
    Description copied from interface: ItemResponseModel
    
    Polytomous item response models may have step parameters. Gets the array of step parameters. Model that do not have step parameters must implement an empty method
    
    Returns:
    
    array of step parameters.
  - setStepStdError
```
public void setStepStdError(double[] stdError)
```
    Description copied from interface: ItemResponseModel
    
    Sets the standard error for the step parameter estimates.
    
    Parameters:
    
    stdError - an array of standard errors for the step parameters.
  - getStepStdError
```
public double[] getStepStdError()
```
    Description copied from interface: ItemResponseModel
    
    Gets that standard errors for each step parameter estimate.
    
    Returns:
    
    an array of step standard errors.
  - getThresholdParameters
```
public double[] getThresholdParameters()
```
    Description copied from interface: ItemResponseModel
    
    Polytomous item response models may use an overall item difficulty parameter and two or more threshold parameters. The difficulty and threshold parameters can be combined to produce the step parameters.
    
    Returns:
    
    array of threshold parameters.
  - getThresholdStdError
```
public double[] getThresholdStdError()
```
    Description copied from interface: ItemResponseModel
    
    Gets the array of standard errors fort eh threshold parameter estimates.
    
    Returns:
    
    array of standard errors.
  - setThresholdStdError
```
public void setThresholdStdError(double[] stdError)
```
    Description copied from interface: ItemResponseModel
    
    Set the threshold standard errors.
    
    Parameters:
    
    stdError - an array of standard errors for the threshold paramter estimates.
  - setStepParameters
```
public void setStepParameters(double[] step)
```
  - setProposalStepParameters
```
public void setProposalStepParameters(double[] step)
```
  - setThresholdParameters
```
public void setThresholdParameters(double[] thresholdParameters)
```
    Description copied from interface: ItemResponseModel
    
    Sets the threshold parameters to particular values. These values will be updated during estimateion unless the item is fixed in ItemResponseModel.setFixed(boolean).
    
    Parameters:
    
    thresholdParameters - array of threshold parameters.
  - setProposalThresholds
```
public void setProposalThresholds(double[] thresholds)
```
    Description copied from interface: ItemResponseModel
    
    Sets the proposed threshold parameters estimates to particular values. These values will be updated during estimation unless the item is fixed in ItemResponseModel.setFixed(boolean).
    
    Parameters:
    
    thresholds - array of proposed threshold parameter estimates.
  - toString
```
public java.lang.String toString()
```
    A string representation of the item parameters. Mainly used for printing and debugging.
    
    Overrides:
    
    toString in class java.lang.Object
    
    Returns:
    
    a string of item parameters.

Class Irm3PL

Field Summary

Fields inherited from class com.itemanalysis.psychometrics.irt.model.AbstractItemResponseModelWithGradient

Constructor Summary

Method Summary

Methods inherited from class com.itemanalysis.psychometrics.irt.model.AbstractItemResponseModelWithGradient

Methods inherited from class java.lang.Object

Constructor Detail

Irm3PL

Irm3PL

Irm3PL

Method Detail

probability

probability

expectedValue

cumulativeProbability

gradient

gradient

hessian

derivTheta2

derivTheta

itemInformationAt

nonZeroPrior

setDiscriminationPrior

setDifficultyPrior

setGuessingPrior

setSlippingPrior

setStepPriorAt

getDiscriminationPrior

getDifficultyPrior

getGuessingPrior

addPriorsToLogLikelihood

addPriorsToLogLikelihoodGradient

incrementMeanSigma

incrementMeanMean

tStarProbability

tSharpProbability

tStarExpectedValue

tSharpExpectedValue

getNumberOfParameters

getNumberOfEstimatedParameters

scale

getItemParameterArray

setStandardErrors

getType

getDifficulty

setDifficulty

getProposalDifficulty

setProposalDifficulty

getDifficultyStdError

setDifficultyStdError

getDiscrimination

setDiscrimination

getProposalDiscrimination

setProposalDiscrimination

getDiscriminationStdError

setDiscriminationStdError

getGuessing

setGuessing

setProposalGuessing

getGuessingStdError

setGuessingStdError

setSlipping

setProposalSlipping

setSlippingStdError

getSlipping

getSlippingStdError

getScalingConstant

acceptAllProposalValues

getStepParameters

setStepStdError

getStepStdError

getThresholdParameters

getThresholdStdError

setThresholdStdError

setStepParameters

setProposalStepParameters

setThresholdParameters

setProposalThresholds

toString