Learning objectives
Aim of the course is the deepening of some statistical models which are of particular relevance in the firm: the design of experiments, the discriminant analysis, Trees with machine learning.
During the course it is also developed a comparison between Machine learning and some statistical models.
Prerequisites
- - -
Course unit content
1. Analysis of variance and design of experiments.
The problem of curvature in the response plan.
Method of the steepest ascent.
Two casual factors model.
Two factors mixed model.
Covariance analysis.
Analysis of variance with stochastic factors.
Part 2. Evaluation of firm with discriminant models.
Fisher Discriminant Analysis.
Quadratic Discriminant Analysis.
ML Discriminant Analysis.
Logistic Discriminat Analysis.
Part 3. Machine learning and the evaluation of firm.
The Nearest Neighbor.
Classification Trees.
Regression Trees.
Random Forests.
Neural Networks.
Full programme
1. Analysis of variance and design of experiments.
The problem of curvature in the response plan with missing values.
Method of the steepest ascent.
Two casual factors model.
Two factors mixed model: one casual and one fixed factor model.
Covariance analysis.
Analysis of variance with stochastic factors.
2. Discriminant Models for the evaluation of firm.
Fisher Discriminant Analysis.
Logistic Discriminat Analysis.
Quadratic Discriminant Analysis.
ML Discriminant Analysis.
Part 3. Cluster Analysis.
Examples of Cluster Analysis with Matlab
Part 4. Principal Components.
Examples of PC with Matlab.
5.Neural Networks.
Bibliography
Montgomery, D. C.(2006)
Design and Analysis of experiments.
McGraw-Hill, New York (Chapters indicated during the course).
Fleury, B. (1997)
A first course in multivariate statistics, Springer, 1997, New York. (Chapters indicated during the course).
Lantz B.(2015) Machine learning with R. Packt Publishing, Birminghan. Open source.(Chapters indicated during the course).
Tibishirami James G.,Witten D., Hastie R. (2013) An Introduction to Statistical Learning with Applications in R, Springer,, New York
Teaching methods
Lectures and laboratory with Matlab.
Assessment methods and criteria
Oral exam.
In the exam the student has to show knowledge of the different models indicated in the programme and developed during the course.
Other information
- - -
