What Resampling is
Resampling is a statistical method used to estimate the properties of an entire population by analyzing a random sample. It involves repeatedly drawing samples from the population and calculating the desired statistic on each sample. The results of the calculations are then combined to estimate the properties of the entire population.
Steps for Resampling:
-
Collect a sample of data from the population.
-
Calculate the statistic of interest on the sample.
-
Repeat steps 1 and 2 a number of times.
-
Record the results of each statistic calculation.
-
Calculate the mean (or other summary statistic) of the results.
-
Compare the mean to the statistic of interest for the population.
-
Interpret the results.
Examples
-
Resampling is commonly used to estimate population parameters from a sample. For example, to estimate the population mean from a given sample, one can use resampling techniques such as bootstrapping or jackknifing.
-
Resampling is also used to compare two populations to determine if their means are significantly different. For example, a researcher could use resampling to compare the average heights of male and female students in a school to determine if there is a statistically significant difference between the two groups.
-
Resampling is used to assess the accuracy of a model. For example, a researcher could use resampling to measure how well a machine learning model predicts a binary outcome (e.g. whether a person will default on a loan or not).