Specification Testing

Identify the parametric form of a model

Apps

Econometric ModelerAnalyze and model econometric time series

Functions

expand all

adftestAugmented Dickey-Fuller test
kpsstestKPSS test for stationarity
lmctestLeybourne-McCabe stationarity test
pptestPhillips-Perron test for one unit root
vratiotestVariance ratio test for random walk
i10testPaired integration and stationarity tests
autocorrSample autocorrelation
parcorrSample partial autocorrelation
crosscorrSample cross-correlation
corrplotPlot variable correlations
lbqtestLjung-Box Q-test for residual autocorrelation
collintestBelsley collinearity diagnostics
gctestBlock-wise Granger causality and block exogeneity tests
archtestEngle test for residual heteroscedasticity
chowtestChow test for structural change
cusumtestCusum test for structural change
recregRecursive linear regression
collintestBelsley collinearity diagnostics
egcitestEngle-Granger cointegration test
jcitestJohansen cointegration test
jcontest Johansen constraint test

Topics

Stationarity

Unit Root Nonstationarity

Learn how to model a unit root process or test for one.

Assess Stationarity of Time Series Using Econometric Modeler

Interactively assess whether a time series is a unit root process using statistical hypothesis tests.

Unit Root Tests

Conduct unit root tests on time series data.

Assess Stationarity of a Time Series

Check whether a linear time series is a unit root process.

Correlation

Implement Box-Jenkins Model Selection and Estimation Using Econometric Modeler App

Interactively implement the Box-Jenkins methodology to select the appropriate number of lags for a conditional mean model. Then, fit the model to data and export the estimated model to the command line to generate forecasts.

Box-Jenkins Methodology

The Box-Jenkins methodology is a five-step process for identifying, selecting, and assessing conditional mean models (for discrete, univariate time series data).

Box-Jenkins Model Selection

Use the Box-Jenkins methodology to select an ARIMA model.

Detect Serial Correlation Using Econometric Modeler App

Interactively assess serial correlation for model specification or Box-Jenkins model selection by plotting the autocorrelation and partial autocorrelation functions (ACF and PACF) and by conducting Ljung-Box Q-tests.

Detect Autocorrelation

Estimate the ACF and PACF, or conduct the Ljung-Box Q-test.

Autocorrelation and Partial Autocorrelation

Autocorrelation and partial autocorrelation measure is the linear dependence of a variable with itself at two points in time.

Ljung-Box Q-Test

The Ljung-Box Q-test is a quantitative way to test for autocorrelation at multiple lags jointly.

Heteroscedasticity

Detect ARCH Effects Using Econometric Modeler App

Interactively assess whether a series has volatility clustering by inspecting correlograms of the squared residuals and by testing for significant ARCH lags.

Detect ARCH Effects

Test for autocorrelation in the squared residuals, or conduct Engle’s ARCH test.

Engle’s ARCH Test

Engle’s ARCH test is a Lagrange multiplier test to assess the significance of ARCH effects.

Structural Change

Check Model Assumptions for Chow Test

Check the model assumptions for a Chow test.

Power of the Chow Test

Estimate the power of a Chow test using a Monte Carlo simulation.

Collinearity

Assess Collinearity Among Multiple Series Using Econometric Modeler App

Interactively assess the strengths and sources of collinearity among multiple series by using Belsley collinearity diagnostics.

Cointegration

Vector Autoregression (VAR) Models

Learn the characteristics of vector autoregression models and how to create them.

Cointegration and Error Correction Analysis

Learn about cointegrated time series and error correction models.

Identifying Single Cointegrating Relations

The Engle-Granger test for cointegration and its limitations.

Identifying Multiple Cointegrating Relations

Learn about the Johansen test for cointegration.

Featured Examples