Bhishan Poudel, Ph.D. Candidate

Data Scientist

Linkedin GitHub twitter stackoverflow



Resampling methods

In statistics there are various ways to get the sample data from the population. The students t-distribution and chi-square distribution are good apporximations for sufficiently large and/or normally-distributed samples. However, when the data is of unknown distribution or sample size is samll resampling tests are required. For example, we can use resamling to get the estimates for sampling distribution of number of connections in a neural net. Here in this blog, I have explored following methods:

(image src: https://www.slideshare.net/hafidztio/resampling-methods)

Resampling

Bootstrapping

Bootstrapping is one of the sampling methods to draw out elements from the population. In bootstrapping method we draw items with replacement. Here, in this blog, I have explored various methods to calculate the bootstrapped sample from a given array. (image src: https://www.wikiwand.com/en/Bootstrapping_(statistics))

Bootstrap




comments powered by Disqus