Quick search:

Who to     contact   Organizing Committees     Topic     Convenors People and contacts List of delegates    Our       sponsors

Conference   venue    Advice for  participants    Useful    information List of deadlines    Other    Brazilian conferences

Registration   details   Accommod-   ation     Local     tours

Conference   theme      Keynote       speakers       *  Main  *       * topics *    Contributed   papers     Posters   Special interest groups Special sessions Administrative meetings

Timetable outline Programme details Monday's programme Tuesday's programme Wednesday's programme Thursday's programme Friday's programme

Notes for participants and authors Guidelines for authors & organizers Announce-   ments

Contributed paper list


   (Wednesday 5th, 11:00-12:30)

How to teach some basic concepts in time series analysis

Authors

Clelia Toloi, Sergio Martins

Presenter

• Clélia Maria de Castro Toloi (Brazil)

Abstract

Time series teaching needs some specific concepts that are not intuitive for most students. In this paper we consider an approach to guide the early stages of time series learning. First we introduce the time series definition and some examples of the sort of series which arise in practice. Here, we comment about their components: trend and/or seasonality. After this we talk about some useful transformations to stabilize the variance, to make the data normally distributed, to eliminate trends and/or seasonal components. These transformations are necessary to reach stationarity, which underlies an important class of stochastic models for describing time series. Finally we introduce the sample autocorrelation function, which measures the linear relationship between observations at different distances apart and provides an idea about a model for generation of the data.