CV_BSADF: Critical values for the backward SADF (BSADF) statistic
In deanfantazzini/bubble: Tests for financial bubbles and explosive behavior

Description Usage Arguments Details Value

This function computes the critical values for the backward SADF (BSADF) statistic by Phillips et al. (2015)

1	CV_BSADF(qe, m, T, r0, address, clust_number, beta = 0.95)

`qe`	is a q x 1 vector of quantiles
`m`	is the number of replications for the simulated DGP
`T`	is the number of simulated time steps
`r0`	is the minimum window size fraction
`address`	is a character object containing the address where the plot density of SADF test stat will be saved in png format
`clust_number`	is the number of clusters for parallel computation
`beta`	is the significance level for the critical values for the BSADF sequence

This function computes the simulated critical values for the backward SADF (BSADF) statistic by Phillips et al. (2015) and saves the plot density of the test statistic as a png file in a specified address. The main difference between the SADF and GSADF tests on one side and BADF BSADF tests on the other side, lies in the data generating process: the SADF and GSADF tests use a random walk with an overall drift value of 1/T, while the the BADF BSADF tests use a random walk with an drift value of 1/r2 for each r2

results is a list containing quantile_bsadf (a q x 1 vector of simulated critical values for the BSADF statistic) and bsadf_cv (a (T-swindow0+1) vector containing the critical values for the BSADF sequence)

deanfantazzini/bubble documentation built on Oct. 22, 2020, 2:43 p.m.