Machine Learning and Predictive Models

Course overview

Course outline

1. 1Day 1: Data Analysis and Simple Regression

   • Fundamentals of Data Analysis Logic
   • Comparing two groups: Means and proportions testing
   • Visualizing group profiles in a single chart
   • Analyzing multiple groups: Means and proportions testing
   • Profiling multiple groups in one chart
   • Introduction to Simple Regression
   • Regression vs. Correlation
   • Sensitivity analysis for quantitative variables
2. 2Day 2: Multiple and Logistic Regressions

   • Overview of Machine Learning principles
   • Understanding Gradient Descent logic
   • Differences between Multiple and Simple Regression
   • Variability analysis in estimations
   • Utilizing dummy variables in models
   • Key distinctions between Logistic and Multiple Regressions
   • Simplifying complex models through Stepwise Regression
3. 3Day 3: Discriminant Analysis

   • Optimized profiling techniques
   • Two-Group Discriminant Function Analysis
   • Case attribution and model evaluation
   • Classification functions and Mahalanobis squared distances
   • Probability-based methods and model reduction
   • Generalized Discriminant Analysis
4. 4Day 4: Decision Trees

   • Introduction to Decision Trees
   • Binary Trees and their quality assessment
   • Rules and techniques for pruning
   • CART Models: Classification and Regression Trees
   • CHAID Trees and Random Forest Trees
5. 5Day 5: Nearest Neighbor, Bayesian, Neural Network and Deep Learning

   • Understanding conditional probabilities for prediction
   • Prediction using probability models
   • Distance-based predictions (Nearest Neighbor)
   • K-Nearest Neighbors methodology
   • Neural Network models: Weights, hidden layers, pros, and cons
   • Introduction to Deep Learning concepts
   • Overview of Big Data principles