Econometrics Toolbox
Model and analyze financial and economic systems using statistical methods
Econometrics Toolbox™ provides functions for modeling economic data. You can select and estimate economic models for simulation and forecasting. For time series modeling and analysis, the toolbox includes univariate Bayesian linear regression, univariate ARIMAX/GARCH composite models with several GARCH variants, multivariate VARX models, and cointegration analysis. It also provides methods for modeling economic systems using state-space models and for estimating using the Kalman filter. You can use a variety of diagnostics for model selection, including hypothesis tests, unit root, stationarity, and structural change.
Get Started:
Time Series Modeling
- Perform modeling tasks, including data preprocessing, data visualization, model identification, and parameter estimations.
- Compare econometric models to ensure the best fit to the data.
- Share results and generate MATLAB code for repeat use.
Econometric Modeler app for time series modeling.
ARIMA
Supported models include AR, MA, ARMA, ARIMA, SARIMA, and ARIMAX.
Importing Time Series Data.
Bayesian Regression
Estimate and simulate Bayesian linear regression models, including Bayesian lasso regression.
Fitting a robust Bayesian linear regression model to data with outliers.
Multivariate Models
Supported models include vector autoregression (VAR) and vector error-correction (VEC)
VAR Model MMSE Forecasts.
Simulate GARCH Model Observations and Conditional Variances.
Markov Chain Models
- Create and simulate discrete-time Markov chains.
- Determine Markov chain asymptotic behavior.
- Compute state redistributions, hitting probabilities, and expected hitting times.
Distribution of states.
State-Space Models
- Create and simulate time-invariant or time-varying state-space models.
- Estimate model parameters from full data sets or from data sets with missing data using the Kalman filter.
The distribution of factors in the Diebold-Li model (a state-space model).
Markov Switching Models
- Analyze multivariate time series data with structural breaks and unobserved latent states.
Simulated responses, innovations, and state indices.
Supported Hypothesis Tests
Perform a variety of pre- and post-estimation diagnostic tests, including:
- Stationarity
- Correlation
- Heteroscedasticity
- Structural change
- Collinearity
- Cointegration
Hypothesis testing.
Markov-Switching Autoregressions
Create a Markov-switching model for analyzing multivariate time series data with structural breaks and unobserved latent states
Discrete-Time Markov Chains
Compute hitting probabilities and expected hitting times
See release notes for details on any of these features and corresponding functions.
Computational Finance Suite
The MATLAB Computational Finance Suite is a set of 12 essential products that enables you to develop quantitative applications for risk management, investment management, econometrics, pricing and valuation, insurance, and algorithmic trading.
