Maximum Likelihood Estimation (MLE) in Normal Distributions: Practical Lab Overview
This lab focuses on verifying Maximum Likelihood Estimation (MLE) as a method to estimate parameters of a normal distribution, specifically mean (μ) and standard deviation (σ). Participants will generate samples from three normal distributions and apply the MLE method to estimate these parameters, ensuring accuracy with the values used in the generation process. Additionally, a Bayesian classifier will employ the estimated parameters to classify input feature values, highlighting the practical applications of MLE. Full coding examples and outputs are provided to facilitate learning.
Maximum Likelihood Estimation (MLE) in Normal Distributions: Practical Lab Overview
E N D
Presentation Transcript
Pattern recognition lab 7 TA : Nouf Al-Harbi :: nouf200@hotmail.com
Lab objective: • Verification of • Maximum Likelihood Estimation MLE
Quick Review MLE • We can estimate the defined parameters of the distribution by using Maximum Likelihood Estimation Method MLE • , in Normal Distribution case • It depends on find the parameters that more representative for the data. • In Normal Distribution ; we concluded these rules to compute parameters :
Verification of MLE Practical Learning 1 • Generate N samples from three normal distributions • Use the MLE method to estimate the parameters for each sample (they should be close to that used in the generation) • Use Bayesian Classifier that uses the estimated parameters to classify any input feature value x
Full code Part 1
Full code Part 2