Predictive Models

1 Lecturer

• Ph.D Carlos Alberto Cardozo Delgado, Full Research Professor - Universidad Manuela Beltran
• email: carlos.cardozo@docentes.umb.edu.co

2 Program

• This is a course of the Master of Biostatistics

3 Some important facts

• It is assumed that you have some experience with statistical concepts, such as, independence, estimators, multiple linear regression model, hypothesis tests.
• In this course we will write programs in the language R. But, if you code in another language, for instance, Python or Matlab, it is also fine.
• Each session have a duration of two hours.

4 General Objective

To build, evaluate and compare predictive statistical learning models.

5 Topics

In this course we will cover the following topics:

5.1 Multiple Linear Regression

5.1.1 The model

5.1.2 Estimating the parameters

5.1.3 Asymptotic properties of the estimators

5.1.4 Testing hypotheses about the parametes

5.1.5 Confidence intervals about the parameters

5.1.6 Prediction interval for a new response value

5.1.7 Model Checking

5.2 K-Nearest Neighbors

5.3 Methods of Validation

5.4 Best Subset of Predictors

5.5 Ridge and Lasso Regression

5.6 Decision Regression Trees

5.7 Logistic Regression

5.8 Discriminant Analysis: Linear and Quadratic

5.9 Decision Clasification Trees

5.10 Ensemble Models: Bagging y Boosting

5.11 Naive Bayes Classifier

5.12 Support Vector Machines

5.13 Neuronal Nets

6 Grading

This course will be grade based on an applied project. The weigths of each stage of the project are:

• 29/02 Formulation of the project 20%
• 28/03 Preliminary report 20%
• 23/05 Final paper 50%
• 30/05 Oral exposition 10%

The Final paper must have a minimum of eigth pages and a maximum of twelve pages.

7 Bibliography

• T. Hastie, R. Tibshirani, J. Friedman (2009).The Elements of Statistical Learning: Data Mining, Inference, and Prediction. Second Edition, Springer Series in Statistics.
• Casella G. and Berger, (2000). Statistical Inference, Second Edition.
• Gareth, J., Witten, D., Hastie, T., and Tibshirani R. (2017). An introduction to Statistical Learning with applications in R, Springer.
• R Core Team, R: A language and environment for statistical computing, R Foundation for Statistical Computing, 2019.
• M. Kuhn, K. Johnson (2013). Applied Predictive Modeling. Springer.
• D. Olson, D. Wu (2020). Predictive Data Mining Models. Second Edition. Springer Verlag.
• C. Bishop (2006). Pattern Recognition and Machine Learning. Springer Verlag.

8 Contact

If you want to know more about this course, please let me know:

• carlos.cardozo@docentes.umb.edu.co