Teaching plan for the course unit

(Short version)


Catalā English Close imatge de maquetació




General information


Course unit name: Linear Models

Course unit code: 361231

Academic year: 2021-2022

Coordinator: Francisco De Asis Carmona Pontaque

Department: Department of Genetics, Microbiology and Statistics

Credits: 6

Single program: S



Estimated learning time

Total number of hours 150


Face-to-face and/or online activities



-  Lecture with practical component

Face-to-face and online




-  Problem-solving class

Face-to-face and online




-  IT-based class

Face-to-face and online



Supervised project


Independent learning




Learning objectives


Referring to knowledge

The main objective of the course is to understand linear models and their application to the most common situations:

— Know the processes of parameter estimation in a linear model.

— Know the decision-making mechanisms associated with the main hypothesis tests in linear models.

— Know how to characterize a simple and multiple linear regression model.

— Know how to validate linear regression models.

— Know how to characterize some simple models of variance analysis.


Referring to abilities, skills

— Know how to solve the parameter estimation of a linear model.

— Know how to analyse the main hypothesis tests in linear models.

— Know how to validate a linear regression model.

— Know how to validate a linear regression model.

— Know how to solve some simple models of variance analysis.

— Be able to interpret the results rigorously.



Teaching blocks


1. Simple linear regression

1.1. Estimation of regression coefficients by least squares

1.2. Decomposition of variability

1.3. Correlation coefficient and determination coefficient

1.4. Inference on regression parameters

1.5. Prediction

1.6. Matrix approach

2. Regression models

2.1. Multiple linear regression

2.2. Measures of best fit

2.3. Inference on regression coefficients

2.4. Standardized regression coefficients

2.5. Polynomial regression

2.6. Introduction to model diagnosis

2.7. Variable selection

2.8. Robust regression

2.9. Simultaneous coefficient penalization

2.10. Logistic regression

3. The linear model

3.1. Estimation of parameters by least squares

3.2. Properties of estimators

3.3. Linear hypothesis testing

3.4. Model testing

3.5. Estimable parametric functions

4. Linear model of variance analysis

4.1. Single factor model

4.2. Comparison of means

4.3. Other models

4.4. Introduction to covariance analysis





Reading and study resources

Consulteu la disponibilitat a CERCABIB


CARMONA, Francesc. Modelos lineales. Barcelona: Publicacions i Edicions de la Universitat de Barcelona, 2005

  A good book on the subject of linear models, but note that it is more advanced in terms of content than is required for this course.

Catāleg UB  Enllaç

FARAWAY, Julian James. Linear Models with R. Chapman & Hall/CRC Press, 2014

Catāleg UB (2n ed. 2015)  Enllaç

FOX, J and WEISBERG, S. An R Companion to Applied Regression. SAGE Publications; Inc, 2018.

PEÑA, Daniel. Estadística: Modelos y Métodos. Vol. 2. Madrid: Alianza, 1991

  This book covers one part of the course.

Catāleg UB  Enllaç

RAWLINGS, John O. Applied Regression Analysis: a research tool. New York [etc.]: Springer, 1998

  This book, in English, is highly recommended for the examples it offers and their corresponding commentaries.

Catāleg UB  Enllaç
Versiķ en línia: Accés restringit als usuaris de la UB, UAB, UPC, UPF, UdG, UdL, URV, UOC, BC, UJI, UVIC  Enllaç

MONTGOMERY, Douglas C. et al. Introduction to Linear Regression Analysis. 2nd ed. New York [etc.]: Wiley, 1992

Catāleg UB  Enllaç
En castellā ed. 2002  Enllaç

OLIVA, Francesc, et al. Propietats i eines d’àlgebra matricial per a estadística. Barcelona : Universitat de Barcelona, 1995

Catāleg UB  Enllaç

Web page

The R Project for Statistical Computing

http://www.r-project.org/  Enllaç