How to Compute Posterior Probabilities

Posterior probability is the recalculated probability of an event occurring after taking new information into consideration. Posterior probabilities are calculated by updating prior probabilities -- the probabilities of events happening based on prior experience -- by using Bayes' Theorem. Named after 18th-century British mathematician Thomas Bayes, the theorem provides a way to revise existing predictions given new or additional evidence. Bayes' Theorem can be used in many applications, such as medicine, finance and economics.To apply Bayes' Theorem, you need an event and you need information.

Instructions

    • 1

      Establish the prior probability of an event. Let Event A be the event that a stock's price rises, and the probability of Event A, or P(A) in Bayes' Theorem, be 50 percent. This is the prior probability of Event A -- the stock's price has proven to rise in 50 percent of the cases studied.

    • 2

      Introduce Event B. Let Event B be the event that interest rates rise, and establish its probability, say 50 percent. Call this P(B) -- it is the probability of Event B happening, and is independent of Event A.

    • 3

      Connect Event A and Event B. Calculate the likelihood, based on new data, that interest rates will rise (Event B) when stock prices rise (Event A). Say that the likelihood -- this is called P(B/A) -- is 20 percent.

    • 4

      Calculate the posterior probability using Bayes' Theorem. The theorem says that the posterior probability P(A/B) equals the likelihood (20 percent) times the prior probability of Event A (50 percent), divided by the prior probability of Event B (50 percent). The posterior probability of stock prices rising when interest rates rise is then (.20 x .50) divided by .50, or 5 percent.

Learnify Hub © www.0685.com All Rights Reserved