Class hierarchy

Static elements shared by the user model, groups and data.

Bellman

Base class for any DP problem and each point \(\theta\) in the endogenous state space.

ExPostSmoothing

Base class for DP problem when choice probabilities are smoothed ex-post.

OneDimensionalChoice

One-dimensional action models with user defined distribution of \(\zeta\).

KeepZ

A OneDimensionalChoice model with discretized approximation to "accepted" past \(\zeta\).

OneStateModel

Base class for a model with a single state.

ExtremeValue

The base class for models that include an additve extreme value error in action value.

McFadden

Myopic choice problem (\(\delta=0.0\)) with standard Extreme Value \(\zeta\).

Rust

Special case of Extreme value.

Normal

The containter class for models that include additve normal smoothing shocks.

NIID

Class for adding correlated normal smoothing shocks to action value.

NnotIID

Class for adding correlated normal smoothing shocks to action value.

Roy

Myopic choice problem (\(\delta=0.0\)) over \(J\) sectors with correlated Normal \(\zeta\).

Group

A Group object stores information for a point \(\gamma\) in the Group Space \(\Gamma\).

Task

Process (span) space or subspace.

Data

Base class for Outcomes and Predictions.

Outcome

A single realization of a discrete DP.

Path

A sequence of outcomes along a single realized DP path.

FPanel

A list of Paths sharing the same fixed effect.

Panel

A heterogenous panel.

OutcomeDataSet

A panel with data-handling tools.

 
   ErgodicOutcomeDataSet
 
An outcome data set for Ergodic DP models with semi-closed form gradients.

Prediction

Predicted distribution across states.

PathPrediction

Predicted outcomes along a path.

PanelPrediction

Store and process path predictions for all fixed effect groups.

PredictionDataSet

Stores data read in as moments and associate them with a panel of predictions.

EndogTrans

Base Task for constructing the transition for endogenous states.

SVTrans

ExTask

Base Task for looping over \(\epsilon\) and \(\eta\).

DumpExogTrans

ExogAuxOut

ExogOutcomes

Compute expected outcomes given exogenous vector values.

ExogUtil

Call Utility().

SemiExTask

Base Task for looping over \(\eta\).

SemiEV

Loop over \(\eta\) to compute EV.

SemiTrans

Loop over \(\eta\) to compute transitions.

GroupTask

The base task for processing \(\gamma\).

CGTask

The task called in CreateSpaces that creates \(\Gamma\).

FETask

The base task for looping over fixed effects.

CCP

Compute Estimate of Conditional Choice Probability from Data.

Method

A container for solution methods.

HotzMiller

Solve a DP model using the Hotz-Miller inverse mapping from conditional choice probabilities.

AguirregabiriaMira

Solve a DP model using the Aguiregabiria Mira iterative prodecure.

ReservationValues

Solve for cutoffs as a non-linear system of equations in a one-dimensional choice problem.

SolveAsSystem

Solve EV as as a non-linear system in a stationary EVExAnte environment.

ValueIteration

Object ot iterate on Bellman's Equation, to solve EV(θ) for all fixed and random effects.

KeaneWolpin

Approximate EV from a subsample of Θ using Keane-Wolpin (1994).

NewtonKantorovich

Iterate on Bellman's equation then switch to N-K iteration.

RETask

The base task for looping over random effects \(\gamma_r\).

DPMixture

RandomEffectsIntegration

Integrate over \(\gamma_r\).

RandomSolve

Loop over random effect values \(\gamma_r\), call GSolve() method for the calling method.

SDTask

UpdateDensity

Dynamically reset the density over random effects given the current fixed effect group.

ThetaTask

Base Class for tasks that loop over the endogenous state space Θ.

CCPspace

CreateTheta

Allocate space for each reachable point \(\theta\) in the state space \(\Theta\).

ReSubSample

Called when subsampling Θ for KeaneWolpin approximation.

Output routines .

The base method for iterating over \(\theta\) during solution methods.

HMGSolve

AMGSolve

KWGSolve

Keane-Wolpin specific version of GSolve.

NKSolve

Replacement for GSsolve used by NewtonKantorovich.

RVGSolve

SaSGSolve

DPSystem

Represent V or R* as a non-linear system.

EVSystem

Holds V as a system of equations.

Rsystem

System of equations for reservation value solutions.

DynamicRsystem

Quantity

Container for discrete variables (DDP) and continuous parameters (FiveO).

AuxiliaryValue

Base Class for variables related to outcomes.

Indicator

An auxiliary value that is an indicator for values other actions or states.

MultiIndicator

Create

Noisy

Add normal noise to a AV() compatible target, either linear or log-linear.

RealizedUtility

Built-in variable that records realized utility.

StaticAux

A wrapper around a function that returns an auxiliary value.

Discrete

Discrete values: actions, states, auxiliaries.

ActionVariable

An element of the action vector \(\alpha\).

BinaryChoice

Easy way to create a binary choice.

StateVariable

Base Class for elements of state vectors, \(\epsilon\), \(\eta\), \(\theta\), \(\gamma\) .

Augmented

State variables that augment another state variable (the base) or otherwise specialize it.

Triggered

Container for augmented state variables in which a value or an action trigger a transition not present in the base state.

ActionTriggered

A state variable that augments a base transition so that the value of an action variable triggers this state to transit to a value.

Forget

When a permanent condition will occur next period because an action is chosen now this state permanently becomes equal to its reset value.

Reset

When the trigger is 1 the transition resets to 0, otherwise it is the base.

RandomTrigger

A state variable that augments a base transition so that with some probability it takes on a special value next period.

UnFreeze

Unfreeze a froze variable under some condition.

ValueTriggered

A state variable that augments a base transition so that the value of a AV()-compatible object triggers this state to transit to a special value.

ForgetAtT

Forget values when t==T (effective next time period).

Freeze

When the trigger value returns TRUE this state freezes at its current value.

SubState

Let a state variable transit only in one sub-period of a Divided clock.

Autonomous

Container for single state variables with a statistically independent transition.

NonRandom

A container class for state variables with a non-random transition.

Accumulator

Add up the values of the target action or state up to a maximum.

ActionAccumulator

Add up the values of a target action up to a maximum.

StateAccumulator

Add up the values of the target state.

Counter

Count periods value(s) of target action or state variable have occurred.

ActionCounter

Track number of periods value(s) of target action variable have occurred.

ActionCounterMaster

Duration

Track number of consecutive periods an action or state variable has had the same value.

StateCounter

Counts periods value(s) of target state s.x have occurred.

StateCounterMaster

Responsible for tracking reachability in a list of state counters for a target.

Deterministic

A state variable with a general non-random transition.

Cycle

Increments each period up to N‾ then returns to 0.

Lagged

Base class for variables that take on previous values of other states or actions.

LaggedAction

Take on the current value of an action variable.

ChoiceAtTbar

Record the value of an action variable at a specified time.

PermanentChoice

Record if an action has ever been non-zero.

LaggedState

Take on the current value of another state variable.

StateAtTbar

Record the value of an state variable q at a specified time, starting the following period.

RetainMatch

Store a new offer or retain its current value.

Tracker

Indicates a state or action took on particular values last period.

ActionTracker

Indicates an action variable took on a value last period.

StateTracker

Indicates another state variable took on a value last period.

PermanentCondition

Indicator for a condition (ever) occuring in the past.

Random

State variables with a non-determinisitic transition.

Absorbing

A binary variable to code an absorbing state.

Asset

Discretized interest-bearing asset.

A variable

A placeholder for variables to be added to the model later.

KeptZeta

A discretized version of a continous ζ.

Markov

A Markov process.

IIDJump

A discrete IID process.

IIDBinary

A binary IID process.

SimpleJump

Equally likely values each period (IID).

 
   Nvariable
 
A jump variable whose acutal values are quantiles of the normal distribution, \(N(\mu,\sigma)\).

 
   MVNvectorized
 
A single IID jump variable which will return a simulated vector of correlated normal variates.

 
   Zvariable
 
A jump variable whose acutal values are quantiles of the standard normal distribution.

 
   Xponential
 
A IID jump variable whose actual values are quantiles of the exponential distribution
with decay rate \(\gamma\).

Jump

A variable that jumps to a new value with probability π, otherwise stays the same.

Tauchen

A one-dimensional correlated discretized normal process using Tauchen's approach.

Rouwenhorst

Rouwenhorst approximation to discretized correlated normal value.

Offer

A Basic Offer Variable.

LogNormalOffer

RandomSpell

A binary random state variable that goes from 1 to 0 with a AV-compatible probability and goes from 0 to 1 based on the value of an action or a CV-compatible object.

RandomUpDown

A state variable that can stay the same, increase by 1 or decrease by 1.

Renewal

Increments randomly based on Pi then returns to 0 if reset=1.

Coevolving

A member of a state block: transition of this variable depends on transition of one or more other state variables.

NormalComponent

Component of a multi-dimensional normal StateBlock.

TimeVariable

Base class for members of a Clock block.

StateBlock

A Block of Coevolving state variables.

Clock

Base class for timing in DP models.

NonStationary

A container for non-stationary clocks.

Aging

Timing in an ordinary finite horizon environment.

StaticP

A static one-shot program (T=1).

Divided

A period is divided into sub-periods.

NonDeterministicAging

Container for non-stationary and non-deterministic aging clocks.

AgeBrackets

Aging within brackets.

Mortality

Deterministic aging with random early death.

 
   Longevity
 
Random mortality and uncertain lifetime.

PhasedTreatment

A sequence of treatment phases with fixed maximum durations.

Regimes

Stationary

Timing in a stationary environment.

Episode

Re-occuring epsiodes with endogenous onset and ending probabilities.

FixedEffectBlock

A Block of FixedEffect group variables.

Regressors

A block of fixed effects that can be created like a vector of demographic variables.

MVIID

Container for a block of discrete multivariate IID variables (can be contemporaneously correlated).

MVNormal

A discrete multivariate normal IID block of contemporaneously correlated variables.

OfferWithLayoff

A combination of an Offer state variable and a status variable, (w,m) .

RandomEffectBlock

A Block of CorrelatedEffect group variables.

TimeInvariant

Base container for an element of the group vector \(\gamma\).

FixedEffect

Base class for a state variable that is non-random and invariant for an individual DP problem.

SubEffect

An element of the FixedEffectBlock.

RandomEffect

Base for that is invariant for an individual DP problem treated as random.

CorrelatedEffect

An element of a RandomEffectBlock.

NormalRandomEffect

A permanent discretize N(0,σ²) random effect.

TauchenRandomEffect

Use Tauchen's method to discretize a normal variable for Fixed Effects.

Parameter

Continuously varying quantity: the base class for parameters of an Objective.

Zauxiliary

A container for auxiliary structures, which helps organize the hierarchy of classes.

CGI

DataColumn

Holds information about a column in the data.

DDPauxiliary

Container for auxiliary classes used in DDP but not elsewhere (directly).

Alpha

Aspects of the Action Space \(A(\theta)\).

Flags

Indicators related to the DP problem.

Hooks

Manages an array (stored in hooks) that holds user-defined static functions/methods to call at different points in the solution process.

Dynamically updated indices into state spaces.

Labels

Contains arrays of labels for variables.

Numbers and sizes of vectors related to the dynamic program.

NKinfo

Newton-Kantorovich iteration information.

Space

Stores information on a set of state variables, such as θ

SubSpace

Stores information on a set of spaces, such as reality or treatment

TrackObj

Contains information on an object (variable, auxiliary outcome, etc) to be tracked.

Discretized

A continuous discretization of a (potentially) continuous quantity.

Equations

Code a system of constraints.

Equality

A container for equality constraints.

InEquality

A container for inequality constraints.

Integration

Container for different integration techniques.

GaussianQuadrature

GQH

Gauss-Hermite Quadrature Integration.

GQL

Gauss-Laguerre Quadrature Integration.

GHK

Smooth Simulation of Multinomial Normal Probabilities and EValues.

LinePoint

Holds one line maximization try.

Point

Store information about a multidimensional point.

CPoint

Store a point for a constrained objective.

MixPoint

Store a point for a mixture objective.

SepPoint

Store a point for a separable objective.

SysPoint

A system point.

Version