Do you know Bradley Efron? He’s a great man.
How did Efron imagine or think about “bootstrap” for the first time?
In his own words:
My first thoughts on the bootstrap centered around variance and bias
estimation. This was natural enough given the bootstrap’s roots in the
jackknife literature, with Quenouille (1949) on bias and Tukey (1958)
on variance setting the agenda. The oldest note I can find says simply
“What is the jackknife an approximation to?” Poor English, but a good
question that resulted in the 1977 Rietz Lecture, “Bootstrap Methods:
Another Look at the Jackknife” (Efron, 1979). Jaeckel’s (1972) Bell
Labs memorandum on the infinitesimal jackknife was particularly
helpful in answering the approximation question.