STAT 111 Lecture Notes - Lecture 12: Sampling Distribution, Confidence Interval, Percentile
Document Summary
Option 1: estimate the se of the statistic by computing sd of bootstrap distribution and then generate a 95% confidence interval by, statistic 2 x se. Option 2: generate p% confidence interval as the using the middle p% of bootstrap statistics. 90 % ci is from 76 to 152. 95% ci is from 73 to 154 to be more confident that you are capturing the true value of the parameter you have to create an interval that captures more values typically use 95% ci. Do not want to trade-off confidence level, want to stay at 95% confidence. Sample size the larger the sample size the more accurate the interval the larger the sample size the smaller the variability in the bootstrap distribution, which will make the interval narrower. The larger the sample size, the more the precise the estimate: decrease the width of the interval.