Formula for calculating bootstrap t-value
Posted: Mon May 15, 2006 3:28 pm
What is the formula used for calculating the bootstrap t-value? I have done a bootstrap with 1000 replicates, and if I use the 1000 estimates for one of the coeffiecients and calculate the t-statistic for the hypothesis that the coefficient=0, (mean divided by stdev divided by root of 1000) then I dont get the same as calculated by SmartPLS. What I am doing here in any case does not make much sense because of the large "bootstrap sample size" the test is based on.