time-series

Time-series unit root and stationarity testing applies ADF checks and trend assessment to determine data integration order. This Skill provides step-by-step stationarity diagnostics, exploratory ACF/PACF plots, and language-specific templates for reproducible workflows.

Building a VAR model for cointegration and Granger causality applies multivariate time-series techniques to capture dynamic relationships. This Skill guides VAR/VECM model selection, cointegration testing, impulse response functions, and residual diagnostics across Python, R, and Stata.

Selecting ARIMA parameters for macroeconomic forecasting applies information criteria and ACF/PACF analysis to identify optimal orders. This Skill provides practical selection criteria, cross-language code templates, and out-of-sample forecast evaluation checks.

Yes, this time-series econometrics workflow supports Python, R, and Stata. It provides modular cross-language code templates and integrated guidance for implementing univariate and multivariate analysis, ensuring reproducible end-to-end workflows.

You need VECM instead of VAR for cointegrated time-series data when unit root tests indicate non-stationary variables share a long-run equilibrium relationship. This Skill guides cointegration testing and VECM modeling to capture both short-run dynamics and long-run adjustment.

Evaluating out-of-sample forecasts for financial time-series models applies residual tests and forecast evaluation checks to assess predictive accuracy. This Skill integrates impulse response functions and out-of-sample validation steps into its end-to-end workflow.