postshock

postshock is a small toolkit for time-series forecasting around shocks/breaks.
It helps you learn from the pre-shock window, forecast into the post-shock period, and optionally adjust those forecasts using information from donor series.

---

Highlights

• Donor Balancing Weights (dbw)
  Compute donor weights on the pre-shock window so donors match the target in covariates. Supports scaling/centering, bounds, and L1/L2 penalties.

• Automatic order selection (auto_garchx)
  Grid search + IC selection for GARCH-X / GJR-GARCH with robust covariance and backcast handling.

• Synthetic prediction (SynthPrediction)
  Fit (S)ARIMA with a post-shock indicator on each donor, aggregate donor “shock effects” via DBW (or uniform weights), and adjust the target’s k-step forecast.

• Synthetic volatility forecasting (SynthVolForecast)
  Forecast variance with a pre-shock GARCH-X on the target and add a donor-based shock adjustment.

• Practical API
  Clean function interfaces, guardrail tests, and documentation for quick adoption.

---

Installation

You can install the development version from GitHub with devtools:

#install.packages("devtools")
devtools::install_github(repo = "DEck13/postshock")