Author: Stephanie Glen<\/p>\n
Resampling<\/a> is a way to reuse data to generate new, hypothetical samples (called resamples<\/em>) that are representative of an underlying population. It’s used when:<\/p>\n Two popular tools are the bootstrap and jackknife. Although they have many similarities (e.g. they both can estimate precision for an estimator \u03b8),\u00a0they do have a few notable differences.<\/p>\n Bootstrapping<\/strong>\u00a0is the most popular resampling method today. It uses sampling with replacement<\/a>\u00a0to estimate the sampling distribution for a desired estimator<\/a>. The main purpose for this particular method is to evaluate the variance of an estimator.<\/strong> It does have many other applications, including:<\/p>\n Bootstrapping has been shown to be an excellent method to estimate many distributions for statistics, sometimes giving better results than traditional normal approximation. It also works well with small samples. It doesn’t perform very well when the model isn’t smooth, is not a good choice for dependent data, missing data, censoring, or data with outliers.<\/p>\n The Jackknife works by sequentially deleting one observation in the data set, then recomputing the desired statistic. It is computationally simpler than bootstrapping, and more orderly (i.e. the procedural steps are the same over and over again). This means that, unlike bootstrapping, it can theoretically be performed by hand. However, it’s still fairly computationally intensive so although in the past it was common to use by-hand calculations, computers are normally used today. One area where it doesn’t perform well for non-smooth statistics (like the median<\/a>) and nonlinear (e.g. the correlation coefficient<\/a>).<\/p>\n The main application for the Jackknife is to reduce bias<\/a>\u00a0and evaluate variance<\/a>\u00a0for an estimator.<\/strong> It can also be used to:<\/p>\n To sum up the differences, Brian Caffo<\/a> offers this great analogy: “As its name suggests, the jackknife is a small, handy tool;\u00a0in contrast to the bootstrap, which is then the moral\u00a0equivalent of a giant workshop full of tools.”<\/em><\/p>\n Some specific differences:<\/p>\n Efron, B. (1982), “The Jackknife, the Bootstrap, and Other Resampling Plans,” SIAM, monograph #38, CBMS-NSF.<\/p>\n Resampling<\/a><\/p>\n Jackknife and Bootstrap<\/a><\/p>\n The Bootstrap and Jackknife<\/a><\/p>\n Bootstrapping, jackknifing and cross validation. Reusing your data<\/a><\/p>\n Evaluation of Jackknife and Bootstrap for Defining Confidence Intervals for Pairwise Agreement Measures<\/a><\/p>\n The Bootstrap and Jackknife Methods for Data Analysis<\/a><\/p>\n Jackknife-Bootstrap<\/a><\/p>\n Descriptive Statistics for Research<\/a><\/p>\n Bootstrap and Jackknife<\/a><\/p>\n<\/div>\n Go to Source<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"\n
What is a Bootstrap?<\/h2>\n
\n
Jackknife<\/h2>\n
\n
Main Differences Between Jackknife and Bootstrap<\/h2>\n
\n
References<\/h2>\n