cvPSYwmboot implements the new bootstrap procedure designed to detect bubbles and crisis periods while mitigating the potential impact of heteroskedasticity and to effect family-wise size control in recursive testing algorithms (Phillips and Shi, forthcoming).

cvPSYwmboot(y, swindow0, IC = 0, adflag = 0, Tb, nboot = 199,
  useParallel = TRUE, nCores)

Arguments

y

A vector. The data.

swindow0

A positive integer. Minimum window size (default = \(T (0.01 + 1.8/\sqrt{T})\), where \(T\) denotes the sample size),

IC

An integer. 0 for fixed lag order (default), 1 for AIC and 2 for BIC (default = 0).

adflag

An integer, lag order when IC=0; maximum number of lags when IC>0 (default = 0).

Tb

A positive integer. The simulated sample size (swindow0+ controlling).

nboot

A positive integer. Number of bootstrap replications (default = 199).

useParallel

Logical. If useParallel=TRUE, use multi core computation.

nCores

A positive integer. Optional. If useParallel=TRUE, the number of cores defaults to all but one.

Value

A matrix. BSADF bootstrap critical value sequence at the 90, 95 and 99 percent level.

References

Phillips, P. C. B., Shi, S., & Yu, J. (2015a). Testing for multiple bubbles: Historical episodes of exuberance and collapse in the S&P 500. International Economic Review, 56(4), 1034--1078.

Phillips, P. C. B., Shi, S., & Yu, J. (2015b). Testing for multiple bubbles: Limit Theory for Real-Time Detectors. International Economic Review, 56(4), 1079--1134.

Phillips, P. C. B., & Shi, S.(forthcoming). Real time monitoring of asset markets: Bubbles and crisis. In Hrishikesh D. Vinod and C.R. Rao (Eds.), Handbook of Statistics Volume 41 - Econometrics Using R.

Examples

y <- rnorm(80) cv <- cvPSYwmboot(y, IC = 0, adflag = 1, Tb = 30, nboot = 99, nCores = 1)