ISMAT 22680

Prediction Mathematical Models

• ApresentaçãoPresentation

  The increasing amount of data available to organizations and its importance for making decisions based on evidence, make it essential to master statistical methods, as well as computational resources that allow them to be implemented. This Curricular Unit (UC) addresses several statistical methods, which allow analyzing dependency and interdependence relationships between economic, environmental, social, marketing variables, among others, and developing predictive models. There are frequent situations in which decision-making depends on the analysis of the evolution of sets of observations made over time (time series), which is why methods of analyzing time series are also addressed in this UC.  
• ProgramaProgramme

  S1: Introduction to data-based forecasting models   S2: Pre-processing and data dimensionality reduction Collection and manipulation of data Principal components and factor analysis   S3: Cluster analysis Hierarchical methods of cluster analysis Non-hierarchical methods of cluster analysis   S4: Predictive models of classification and regression Linear and logistic regression Classification and regression trees Comparison of performance of predictive models   S5: Time Series Notation and nomenclature Trend and seasonality Moving averages and exponential smoothing Identification, estimation, diagnosis and prediction with ARIMA and SARIMA models   S6: Business Management applications using data analysis softwares (R and jamovi).  
• ObjectivosObjectives

  At the end of this Curricular Unit, students should be able to: LO1: Identify appropriate techniques for making data-based predictions; LO2: Create, manipulate and reduce the dimensionality of data; LO3: Apply cluster analysis techniques; LO4: Develop and compare predictive models of regression and classification; LO5: Model time series and use the models for predictive purposes; LO6: Use softwares (R and jamovi) for model development and forecasting; LO7: Critically analyze the predictions obtained, taking into account the context of the problems.  
• BibliografiaBibliography

  Cowpertwait, P. & Metcalfe, A. (2009). Introductory time series with R . New York: Springer. Maindonald, J. & Braun, W. J. (2010). Data analysis and graphics using R: an example-based approach. (3rd ed.). United Kingdom: Cambridge University Press. Tabachnick, B. & Fidell, L. (2012) Using multivariate statistics . (6ª ed.) Boston: Pearson Education. Tufféry, S. (2011). Data mining and statistics for decision making . United Kingdom: John Wiley & Sons Ltd.    
• MetodologiaMethodology

  The teaching methodology includes the expository method (TM1) to present the contents, the demonstrative method (TM2) to illustrate its application to practical cases and the active method (TM3) to solve classroom exercises with the use of a computer. Classes are complemented with videos and other digital resources, made by the professor, to support the teaching-learning process.  The evaluation is made by continuous assessment or written exam. The continuous assessment consists of: - Exercises (Moodle): 20%; - Group work: 40%; - Written test: 40%
• LínguaLanguage

  Português
• TipoType

  Semestral
• ECTS

  5
• NaturezaNature

  Mandatory
• EstágioInternship

  Não