Bootstrap process for estimating the standard error of statistic s(x). B bootstrap sample are generatied from original data. Finally the standard deviation of the values s(x1),s(x2)..s(xB) is our estimate of the standard error of s(x). The bootstrap estimate of standard error is the standard deviation of bootstrap replications. Typical value for B, number of bootstrap samples range from 50 to 200 for stand.error estimation
What I understand is, that only when it is:
- replicate and random
thats called bagging. Does it mean, when you just/only do bootstrap without an aggregation at the end you only use the standard deviation of the bootstraps replications togheter, than you call it bootstrap.
If you do the same, but at the end you take an aggregation of voting or anything else, than you call it bagging.
am I right?
So Breiman dont gives an exactly answer of the aggregation method in the end. But his method (voting) is different of the original bootstrap method (standard deviation) in Tibshiranis book which refers to Efron´s original (Efron´s method).
Can we say that this is a different of bagging and bootstrap aggregating? Of course roughly said that´s the same. (But im looking for the correct answer).