Simulate the finite sample critical values for the PSY test.

cvPSYmc implements the real time bubble detection procedure of Phillips, Shi and Yu (2015a,b)

cvPSYmc(obs, swindow0, IC = 0, adflag = 0, nrep = 199,
  multiplicity = TRUE, Tb, useParallel = TRUE, nCores)

Arguments

obs	A positive integer. The number of observations.
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).
nrep	A positive integer. Number of replications (default = 199).
multiplicity	Logical. If multiplicity=TRUE, use family-wise size control in the recursive testing algorithms.
Tb	A positive integer. The simulated sample size (swindow0+ controlling). Ignored if multiplicity=FALSE.
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.

Examples

cv <- cvPSYmc(80, IC = 0, adflag = 1, Tb = 30, nrep = 99, nCores = 1)