Time Series MT (TSMT)#

A time series package for GAUSS.

Description#

Provides tools for comprehensive treatment of time series models, including model diagnostics, MLE and state-space estimation, and forecasts. Time Series MT also includes tools for managing panel series data and estimating and diagnosing panel series models, including random effects and fixed effects.

Installation#

Please contact us with to request pricing and installation information.

If you already own TSMT, you can use the GAUSS Package Manager to install TSMT.

Requires GAUSS/GAUSS Engine v18 or higher.

Commands#

Univariate Time Series Models#

Conditional mean models#

`arimafit()`	Estimates coefficients of a univariate time series model with autoregressive-moving average errors. Model may include fixed regressors.
`arimass()`	Estimates ARIMA models using a state space representation, the Kalman filter, and maximum likelihood.
`arimapredict()`	Estimates forecasts using estimation results obtained from `arimaFit()`.
`autoregfit()`	Estimates coefficients of a regression model with autoregressive errors of any specified order.
`sarimass()`	Estimates SARIMA models using a state space representation, the Kalman filter, and maximum likelihood.

Conditional variance models#

`garchfit()`	Estimates univariate GARCH model.
`garchgjrfit()`	Estimates GARCH-GJR model.
`garchmfit()`	Estimates GARCH-in-mean model.
`igarchfit()`	Estimates integrated GARCH model, i.e., a model containing a unit root.

Multivariate Time Series Models#

Conditional mean models#

`varmafit()`	Computes exact maximum likelihood parameter estimates for a VARMA model.
`ecmfit()`	Calculate and return parameter estimates for an error correction model.
`svarfit()`	Estimate structural VAR models using short-run, long-run, or sign restrictions.

Panel data and other models#

`tscsfit()`	Estimates the parameters of the pooled time-series cross-section regression model.
`lsdvfit()`	Estimates coefficients of a regression model with autoregressive errors of any specified order.
`kalmanfilter()`	Data filtering algorithm.

Nonlinear Time Series Model#

`switchfit()`	Estimates the parameters of the Markov switching regression model.
`sbreak()`	Estimates the m-break structural break model.
`tartest()`	Estimates the p^th order threshold autoregression model.

Miscellaneous#

`aggdata()`	Aggregates time series data from higher to lower frequency.
`breitung()`	Panel series unit root testing.
`ips()`	Conduct the Im, Pesaran, and Shin panel data unit root test.
`cdtest()`	Runs cross-sectional dependence, CD, tests for panel data.
`dfgls()`	Test for unit root in univariate time series.
`hansen()`	Test for stability of all parameters.
`httest()`	Perform the Harris–Tzavalis panel series unit root testing.
`kpss()`	Test for stationarity using a Lagrange Multiplier score statistic.
`lagreport()`	Compute and graph the autocorrelation function and partial autocorrelation function for a time series.
`rolling()`	Performs rolling OLS regressions for a provided vector of dependent data and matrix of independent regressors.
`selectlags()`	Select lags based on method of statistical inference.
`startest()`	Estimates a p^th order threshold autoregression and tests the hypothesis of a linear autoregression, using the statistics described in “Inference when a nuisance parameter is not identified under the null hypothesis.” (Hansen, 1996).
`tsdiff()`	Differences matrices with or without seasonality.
`varmapredict()`	Calculates forecasts from a VARMAX model.
`vmdetrendmt()`	Detrends data. (DEPRECATED)
`vmdiffmt()`	Differences matrices.
`vmsdetrend()`	Seasonally detrends data.
`vmsdiffmt()`	Seasonally differences matrices. (DEPRECATED)
`zandrews()`	The Zivot and Andrews (1992) unit root test uses a t-test statistic for testing the null hypothesis of stationarity.

Further Reading#