estimates the parametric process proc from data.


estimates the parameters p, q, with starting values p0, q0, .


estimates process proc with starting values taken from the instantiated process iproc.

Details and Options

  • EstimatedProcess returns the symbolic process proc with parameter estimates inserted for any non-numeric values.
  • The data can be given in the following forms:
  • {s0,}a path with state si at time i
    {{t0,s0},}a path with state si at time ti
    TemporalData[]one or several paths
  • The times ti and states si must belong to the time and state domain of the process proc.
  • The process proc can be any parametric scalar- or vector-valued process.
  • The following options can be given:
  • AccuracyGoalAutomaticthe accuracy sought
    ProcessEstimatorAutomaticwhat process parameter estimator to use
    PrecisionGoalAutomaticthe precision sought
    WorkingPrecisionAutomaticthe precision used in internal computations
  • The following basic settings can be used for ProcessEstimator:
  • Automaticautomatically choose the parameter estimator
    "MaximumLikelihood"maximize the log likelihood directly
    "MethodOfMoments"match covariance
  • Special settings for ProcessEstimator are documented under the individual random process reference pages.
  • Additional settings for time series processes include "MaximumConditionalLikelihood" and "SpectralEstimator".
  • Additional settings for HiddenMarkovProcess include "BaumWelch" and "ViterbiTraining".


open allclose all

Basic Examples  (2)

Estimate the parameter of a PoissonProcess:

Compare simulations of the estimated process to the original data:

Find parameters for an ARProcess:

Compare correlation functions for data and the estimated process:

Scope  (9)

Parametric Processes  (3)

Estimate the parameters for a RandomWalkProcess:

Estimate the parameters for a RenewalProcess:

Estimate the parameters for a Wiener process:

Time Series Processes  (3)

Estimate the parameters for an ARProcess:

Compare covariance functions:

Estimate an ARMAProcess:

Compare correlation functions:

Provide initial values for the estimation of an ARProcess:

Solve for repeated parameters:

Queueing Processes  (1)

Parameter estimation for an M/M/1 queue:

Use the corresponding random path for the data:

Compare system sizes for the original and estimated processes:

Finite Markov Processes  (2)

Estimate a four-state discrete Markov process:

Estimate a four-state continuous Markov process:

Options  (5)

WorkingPrecision  (1)

Estimate a process using default machine precision:

Use higher precision:

ProcessEstimator  (4)

Maximum likelihood for a parametric process:

Time series process:

Method of moments for a parametric process:

Time series process:

Maximize conditional likelihood for a time series process:

Spectral estimator for a time series process:

Applications  (2)

Model the mean daily temperature for Champaign in August 2012:

Find process parameters:

Compare CorrelationFunction of the model and the data:

Forecast the daily exchange rates of the euro to the dollar from May 2012 through September 2012:

Fit an AR process to the exchange rates:

Forecast for 20 business days ahead:

Plot the forecast with original data:

Properties & Relations  (2)

EstimatedProcess estimates a parametric process:

FindProcessParameters returns a list of parameter estimates for the process:

EstimatedProcess estimates a parametric process:

EstimatedDistribution estimates a parametric distribution:

Introduced in 2012