Frequentist and Bayesian approaches to interpreting probability and making decisions based on data

Frequentist v/s Baysian

Frequentist and Bayesian approaches to interpreting probability and making decisions based on data
Photo by Robert Stump on Unsplash

Introduction

Frequentist and Bayesian approaches are two different ways to interpret probability and make decisions based on data.

In frequentist statistics, the probability is used to describe the likelihood of an event occurring in the long run, based on the frequency of past events. For example, if you flip a coin 10 times and it lands heads 7 times, the frequentist probability of the coin landing heads on the next flip is 7/10, or 70%.

On the other hand, in Bayesian statistics, the probability is used to describe the degree of belief that an event will occur. This belief can be based on both past data and personal judgment. In the coin flipping example, a Bayesian might use past data (e.g. the coin landing heads 7 out of 10 times) as well as their own personal beliefs (e.g. the coin is fair) to estimate the probability of the coin landing heads on the next flip.

So, the main difference between frequentist and Bayesian approaches is that frequentist probability is based on the frequency of past events, while Bayesian probability is based on both past data and personal judgment.

Here are a few more examples to get a clear understanding.

Example 1:

Imagine you are trying to estimate the probability that a certain medical treatment will be effective in reducing blood pressure. In a frequentist approach, you would conduct a randomized controlled trial in which you give the treatment to a group of patients and measure their blood pressure before and after the treatment. You would then calculate the percentage of patients who experienced a reduction in blood pressure and use this as the probability of the treatment being effective.

In a Bayesian approach, you might also use data from the randomized controlled trial to estimate the probability of the treatment being effective. However, you might also incorporate other factors, such as your own expertise and knowledge about similar treatments, to revise your probability estimate.

Example 2:

Imagine you are trying to estimate the probability that a certain stock will go up in value over the next year. In a frequentist approach, you might look at the historical data on the stock’s performance and calculate the percentage of times it has gone up in the past.

In a Bayesian approach, you might also consider the historical data on the stock’s performance, but you might also incorporate other factors, such as the overall performance of the stock market, the performance of similar stocks, and your own expert judgment about the company’s future prospects.


Thanks for reading.

Subscribe to CodeCraft by Anyesh | Programming, ML & AI Tutorials

Don’t miss out on the latest issues. Sign up now to get access to the library of members-only issues.
jamie@example.com
Subscribe