Machine Learning with Logistic Regression in Excel, R, and Power BI

Introduction

• Apply logistic regressions to solve problems
• What you should know
• Introduction to the course project
• Configuring the Excel Solver Add-in
• Working with R
• Configuring R in Power BI

1. Distributions and Probabilities

• Introducing AI and logistic regression
• Differentiating between odds and probabilities
• Differentiating between distributions
• Calculating logs and exponents
• Sigmoid curve
• Utilizing training and testing data sets

2. Binomial Logistic Regression

• Calculating linear regression
• Working with the logit model
• Calculating log likelihood
• Constructing MLE
• Solving MLE
• Predicting outcomes
• Visualizing logistic regression
• Challenge: Calculating logistic regression
• Solution: Calculating logistic regression

3. Fine-Tuning the Model

• Adding more independent variables
• Transforming variables
• Calculating correlations
• Using statistics
• Configuring confusion tables
• Challenge: Fine-tuning the model
• Solution: Fine-tuning the model

4. Multinomial Regression

• Calculating odds for multinomial models
• Calculating probabilities for multinomial models
• Calculating multinomial log likelihoods
• Running MLE
• Making the predictions

5. Working in Power BI with R

• Running R scripts in the Power Query Editor
• Running R standard visuals
• Interacting between visual components
• Challenge: Moving into Power BI
• Solution: Moving into Power BI

Conclusion

• Next steps with logistic regressions