Minimizing post-shock forecasting error through aggregation of outside information

We develop a forecasting methodology for providing credible forecasts time series that have recently undergone shock. achieve this by borrowing knowledge from other similar shocks which post-shock outcomes are observed. Three shock effect estimators motivated with the aim of minimizing average forecast risk. propose risk-reduction propositions provide conditions establish when our works. Bootstrap and leave-one-out cross-validation procedures provided to prospectively assess performance methodology. Several simulated data examples two real Conoco Phillips Apple stock price verification illustration.