02QUBRS

A.A. 2020/21

Course Language

Inglese

Course degree

Doctorate Research in Urban And Regional Development - Torino

Course structure

Teaching | Hours |
---|---|

Lezioni | 15 |

Teachers

Teacher | Status | SSD | h.Les | h.Ex | h.Lab | h.Tut | Years teaching |
---|---|---|---|---|---|---|---|

Bellone Tamara | 15 | 0 | 0 | 0 | 8 |

Teaching assistant

Context

SSD | CFU | Activities | Area context |
---|---|---|---|

*** N/A *** |

The huge amount of data supplied by advanced technologies, requires a much needed treatment by new processes. So, an improved usage of classic methods, already learned in previous studies, is to be encouraged.
When the investigated phenomenon is to be considered from a number of different viewpoints, it is hard the proper choice of a method for analysis. The said choice is linked to the scope of analysis: the multivariate analysis is the collective name for a melting pot of techniques: reduction and organization of the data, search of dependence among variables, statistical inferences.
The course is intended a san offer of general ideas and methods at practicable level for people already in possess of basic knowledge of Statistics. Also, the course hopes to make immediately usable the explained methods as soon as possible.

The huge amount of data supplied by advanced technologies, requires a much needed treatment by new processes. So, an improved usage of classic methods, already learned in previous studies, is to be encouraged.
When the investigated phenomenon is to be considered from a number of different viewpoints, it is hard the proper choice of a method for analysis. The said choice is linked to the scope of analysis: the multivariate analysis is the collective name for a melting pot of techniques: reduction and organization of the data, search of dependence among variables, statistical inferences.
The course is intended a san offer of general ideas and methods at practicable level for people already in possess of basic knowledge of Statistics. Also, the course hopes to make immediately usable the explained methods as soon as possible.

Some basic knowledge of elementary statistics is suitable, however, the lessons are open to everybody.

Some basic knowledge of elementary statistics is suitable, however, the lessons are open to everybody.

Aspects of Multivariate Analysis and applications of Multivariate Techniques. Examples in the field of Engineering Natural Sciences and Human Sciences
The organization and representation of the data. The geometry of the sample. The multivariate normal distribution. Inferences about a mean vector. Hotelling’s T2 test. Confidence regions and simultaneous comparisons of components means. Comparing mean vectors from two populations.
Comparing mean vectors from several multivariate populations: Multivariate Analysis of Variance (MANOVA methods: one way, two way, one way with interaction, ...). Multivariate Analysis of Variance-Covariance (MANCOVA).
The classic solution: multiple regression model. Inferences about the regression model. Multivariate multiple regression
Data reduction and interpretation: Principal Components Analysis and Factor Analysis. Comparing Principal Components Analysis and Ridge Regression, Partial Least Square and Total Least Square. Comparing Factor Analysis and Principal Component Analysis.
Discrimination and classification: Fisher’s discriminant function to separate two or several populations. Comparison with Logistic regression.

Aspects of Multivariate Analysis and applications of Multivariate Techniques. Examples in the field of Engineering Natural Sciences and Human Sciences
The organization and representation of the data. The geometry of the sample. The multivariate normal distribution. Inferences about a mean vector. Hotelling’s T2 test. Confidence regions and simultaneous comparisons of components means. Comparing mean vectors from two populations.
Comparing mean vectors from several multivariate populations: Multivariate Analysis of Variance (MANOVA methods: one way, two way, one way with interaction, ...). Multivariate Analysis of Variance-Covariance (MANCOVA).
The classic solution: multiple regression model. Inferences about the regression model. Multivariate multiple regression
Data reduction and interpretation: Principal Components Analysis and Factor Analysis. Comparing Principal Components Analysis and Ridge Regression, Partial Least Square and Total Least Square. Comparing Factor Analysis and Principal Component Analysis.
Discrimination and classification: Fisher’s discriminant function to separate two or several populations. Comparison with Logistic regression.

In presenza

On site

Presentazione orale

Oral presentation

P.D.1-1 - Novembre

P.D.1-1 - November

© Politecnico di Torino

Corso Duca degli Abruzzi, 24 - 10129 Torino, ITALY

Corso Duca degli Abruzzi, 24 - 10129 Torino, ITALY