NRMstatsML News

NRMstatsML 0.1.0 (initial release)

New features

• trendML module: nrm_trend(), nrm_mann_kendall(), nrm_sens_slope(), nrm_structural_break() — non-parametric trend detection and Bai-Perron structural break analysis.

• multiSysML module: nrm_multivariate(), nrm_pls(), nrm_sem() — scaled OLS, Partial Least Squares, and Structural Equation Modelling.

• responseML module: nrm_response_curve() (quadratic, linear, and Mitscherlich types) and nrm_optimize_input() for economic optimum calculation.

• tsML module: nrm_arima() and nrm_forecast() wrapping forecast::auto.arima() with 95 % prediction intervals.

• panelML module: nrm_panel() (fixed/random effects with Hausman test) and nrm_did() (Difference-in-Differences).

• uncertaintyML module: nrm_bootstrap(), nrm_monte_carlo(), and unified nrm_uncertainty() dispatcher.

• autoML module: nrm_automl() for automated cross-validated model selection and nrm_benchmark() for hold-out evaluation.

• Generic helpers: nrm_data_check(), nrm_summary(), nrm_plot().

• Example dataset nrm_example: 20-year synthetic NRM time series.

• Vignette: Getting Started with NRMstatsML.