This repository contains the work completed as part of the "Applied Probabilistic Models" course. The practical sessions aimed to provide a comprehensive understanding of Bayesian inference and probabilistic modeling through hands-on exercises and applications. The focus was on implementing theoretical concepts in R and applying them to solve real-world problems. These labs were made in groups of 3.
This session focused on simulating joint distributions and analyzing conditional probabilities. Techniques such as rejection sampling and Gibbs sampling were implemented. Linear regression was applied to simulated data to explore relationships between variables.
In this lab, the objective was to develop probabilistic models to detect change points in binary sequences. Statistical inference methods were employed to analyze signal transitions and determine the change points accurately (Gibbs sampling).
This session explored Bayesian methods for computing posterior distributions and predictive probabilities. The focus was on applying these techniques to make informed decisions under uncertainty in various contexts. Use of Metropolis-Hastings algorithm and Gibbs sampling.
- Programming Language: R
- Topics Covered: Bayesian inference, probabilistic models, sampling methods, regression analysis, change point detection.