Print SVAR Identification Steps — fSVAR

Prints a numbered guide to the proxy-SVAR identification and invertibility-testing procedure to the R console.

Usage

fSVAR_steps()

Value

Called for its side-effect of printing to the console. Returns NULL invisibly.

Details

Step 1: Estimate the VAR(p) and extract the residuals.

Step 2: Regress the instrument on the current residuals plus f leads; save the fitted value.

Step 3: F-test whether the lead coefficients are jointly zero. Rejection implies the shock is non-invertible.

Step 4: If invertible, apply the standard Stock-Watson IV procedure to obtain IRFs and unit-variance shocks.

Step 5: If non-invertible, test recoverability via the Ljung-Box test on the fitted instrument series.

Step 6: If recoverable (null of no autocorrelation not rejected), repeat the IV procedure adding residual leads to the first-stage regression.

Step 7: If recoverability is rejected, SVAR analysis with this instrument is not feasible.

Examples

fSVAR_steps()
#> Step 1: Estimate the VAR(p) and extract the residuals.
#> Step 2: Regress the instrument on current residuals plus f leads. Save the fitted value.
#> Step 3: F-test that lead coefficients are jointly zero. Rejection implies non-invertibility.
#> Step 4: If invertible, apply the Stock-Watson IV procedure for IRFs and unit-variance shocks.
#> Step 5: If non-invertible, test recoverability via the Ljung-Box test on the fitted instrument.
#> Step 6: If recoverable, redo the IV procedure adding residual leads to the first-stage regression.
#> Step 7: If recoverability is rejected, SVAR analysis with this instrument is not feasible.