49513 - Statistics for Business

Learning outcomes

The aim of the course is to deliver to the student the basics of data mining with special attention toward business and marketing decisions. Focus will be given on the critical interpretation and policy implications of the results of regression analysis, classification and clustering. The software environment which will be used for data analysis and examples is R.

Course contents

1. Statistical inference: confidence intervals, hypothesis testing. Introduction to R.

2. Multiple regression model and its extensions: logistic regression for binary data. Regression trees. R examples.

3. Model selection, diagnostics and residual analysis.

4. Classification methods. Discriminant analysis. R examples.

5. Clustering methods. Supervised and unsupervised learning. R examples.

Readings/Bibliography

• Borra S., Di Ciaccio A., STATISTICA, Metodologie per le scienze economiche e sociali, McGraw-Hill, 2004. http://www.ateneonline.it/diciaccio/

Chap.11 § except 11.7;

Chap.12 § 12.1-12.4;

Chap.13 § 13.1-13.6, 13.9;

Chap.14 § 14.1-14.2 (up to 14.2.3);

Chap.17 § 17.5-17.7;

Chap.19 (http://www.ateneonline.it/diciaccio/capitolo_19.pdf), 19.1-19.7;

• Azzalini A., Scarpa B., Analisi dei dati e data mining, Springer, 2004.

  Cap.1 §

  Cap.2 § 2.1 (except 2.1.3), 2.3 ( except 2.3.3), 2.4;

  Cap 3 § except 3.5.4;

  Cap 5 § 5.1, 5.2;

  Cap 6 § to be defined;

• A. Montanari: Appunti sulla Regressione Multiplahttp://www2.stat.unibo.it/montanari/Didattica/dispensa2.pdf

  § 2.10

Further readings:

• S. Iacus, G. Masarotto, Laboratorio di Statistica con R, McGraw-Hill. 2003.

Teaching methods

Lectures and Classes.

Assessment methods

Oral examination.

Teaching tools

Links to further information

http://www2.stat.unibo.it/giannerini/

Office hours

See the website of Simone Giannerini