The bootstrap
The bootstrap is the most popular resampling method to express the uncertainty of an estimate; in other words, to estimate the variance of an estimated statistic of interest. But why is it called bootstrap? Tall boots may have a tab, loop or handle at the top known as a bootstrap; see Figure 7.1:
This bootstrap allows us to use our fingers to pull the boots on. But the term is used as a synonym for more. In the 19th century, the idiom "to pull oneself up by one's bootstraps was already being used as an example of an impossible task:
"It is conjectured that Mr. Murphee will now be enabled to hand himself over the Cumberland river or a barn yard fence by the straps of his boots" (Freeman 2009). This is just what the bootstrap is about in statistics. We will see that we use a bootstrap to make inference just with our boots (sample data).
In the following section, we will show with a motivating example that we get basically the same results with the bootstrap in comparison to analytical...
