Objectifs

Important examples of time series data include monthly number of claims filed from an insurance portfolio, cash-flows over time, daily climate data (temperature, snowfall, rainfall), the number of sun spots observed etc. Insurance and finance applications aim at explaining the observed time series data in order to understand the underlying phenomenon which can lead to time series models for future predictions. Typically, time series data contain information related to trends, business cycles, sesonality being further subject to residual errors.

This course shall discuss both probabilistic & statistical aspects of stationary time series. Basics of non-stationary & multivariate times series will be also discussed. R will be the main platform for the illustration of various concepts and the statistical analysis of real/simulated time series data.

Contenus

• Introduction to ARMA models
• Probabilistic background
• ARIMA models
• Models for non-stationary times series
• Financial time series
• Estimation & prediction
• Multivariate time series

The course will be accompanied by exercises and facultative R programming tasks.

Références

Brockwell, P.J., Davis, R.A. (2002) Introduction to time series and forecasting. Second Edition, Springer
Ruppert, D., Matteson, D.S. (2015) Statistics and data analysis for financial engineering, with R examples. Second Edition, Springer
Shumway, R.H., Stoffer, D.S. (2011) Time series analysis and its applications, with R examples. Third Edition, Springer

Pré-requis

Basics in probability & statistics

Evaluation

1ère tentative

Examen:

Ecrit 2h00 heures

Documentation:

Non autorisée

Calculatrice:

Non autorisée

Evaluation:

Final Grade = Grade of Final Exam*

*Exam conditions may be subject to changes depending on the pandemic situation.

Rattrapage

Examen:

Ecrit 2h00 heures

Documentation:

Non autorisée

Calculatrice:

Non autorisée

Evaluation:

Final Grade = Grade of Final Exam*

*Exam conditions may be subject to changes depending on the pandemic situation.