microsim.statistics.regression.GeneralisedOrderedRegression<E1>

public class GeneralisedOrderedRegression<E1 extends Enum<E1> & IntegerValuedEnum> extends Object implements IDiscreteChoiceModel

Generalised Ordered Discrete Variable Models

Author:: Justin van de Ven Let y define a discrete valued set {y_0,y_1,..., y_n} Each y_i is an indicator variable for discrete value i, where higher values of i reflect some natural ordering of the set Define yhat_j = sum(y_j+1,...,y_n) yhatstar_j = Xb_j - e_j yhat_j = 1 if yhatstar_j>=0 and 0 otherwise P(yhat_j=1|X) = P(yhatstar_j>=0|X) = P(Xb_j-e_j>=0) = F(Xb_j)

Constructor Details
- GeneralisedOrderedRegression
  
  public GeneralisedOrderedRegression(RegressionType type, Class<E1> enumType, MultiKeyCoefficientMap multinomialCoefficients)
Method Details
- getEventList
  
  public List<E1> getEventList()
  
  Specified by:
  
  getEventList in interface IDiscreteChoiceModel
- getProbability
  
  public <E extends Enum<E> & IntegerValuedEnum, E2 extends Enum<E2>> double getProbability(E event, IDoubleSource iDblSrc, Class<E2> Regressors)
  
  Specified by:
  
  getProbability in interface IDiscreteChoiceModel
- getProbabilities
  
  public <E extends Enum<E> & IntegerValuedEnum, E2 extends Enum<E2>> Map<E,Double> getProbabilities(IDoubleSource iDblSrc, Class<E2> Regressors)
  
  Specified by:
  
  getProbabilities in interface IDiscreteChoiceModel

Class GeneralisedOrderedRegression<E1 extends Enum<E1> & IntegerValuedEnum>

Constructor Summary