LEA PETRELLA
Structure:
Dipartimento di METODI E MODELLI PER L'ECONOMIA, IL TERRITORIO, LA FINANZA
SSD:
STAT-01/A
Profilo Research

Notizie

Avvisi Generali del Docente

 

L'esame di Analisi delle Serie Storiche e di Modelli per l'Analisi dei Flussi Turistici si svolgerà il giorno 28-01-2025 alle ore 10.30 in aula Di Fresco.

 

TIME SERIES AND FINANCIAL TIME SERIES

Syllabus:

  • Introduction to time series 
  • Graphical analysis
  • The components: trend, seasonality and error term.
  • How to eliminate the trend the seasonality
  • The concepts of stationarity and inveribility.
  • The autocorrelation function
  • ARMA models: The AR(p), the MA(q), the ARMA(p,q), the ARIMA(p,d,q): methodology and properties
  • Estimation of parameters: a likelihood approach
  • How to chose an ARMA model: a likelihood approach.
  • Read data analysis using R
  • Stylized facts of financial data
  • ARCH models, methodological aspects and properties
  • GARCH models, mehodological aspects and properties
  • Generalization of GARCH models.
  • During classes we will use the software R

Teaching material will be provided

Materiale didattico: 

 

ANALISI DELLE SERIE STORICHE

Programma del corso:

  • Introduzione al concetto di serie storica.
  • Analisi grafica.
  • Le componenti di trend e di stagionalita'.
  • destagionalizzazione
  • Il concetto di stazionarietà ed invertibilita' .
  • La funzione di autocovarianza
  • Esempi di processi non stazionari e non invertibili
  • Modelli ARMA: i modelli AR(p), MA(q), ARMA(p,q), ARIMA(p,d,q); caratteristiche e proprieta'.
  • Individuazione dei modelli, stima dei parametri e previsione con l'utilizzo di R
  • Criteri di scelta dei modelli.
  • Caratteristiche dei dati finanziari
  • Introduzione ai modelli ARCH e GARCH
  • Durante tutto il corso si fara' uso del software R

 

Materiale didattico: fornito dal docente

 

Collegamenti utili per trovare serie storiche:

Materiale didattico: 

 

METODI STATISTICI AVANZATI

 

Obiettivi del corso:

Il corso intende fornire agli studenti gli strumenti teorici e pratici adeguati per modellare fenomeni reali. Tutte le analisi teoriche verranno supportate dall’utilizzo del pacchetto statistico R in laboratorio informatico.

Programma:

Richiami di probabilità. Distribuzioni di variabili aleatorie discrete e continue: Bernoulli, Binomiale Poisson, Ipergeometrica, Uniforme, Normale, Chi quadrato, Gamma, Beta, t di Student, F di Fisher. Simulazione e analisi attraverso il pacchetto statistico R. Approccio inferenziale basato sulla verosimiglianza, stima verifica delle ipotesi ed intervalli di confidenza. Utilizzo del pacchetto statistico R. Modello di regressione lineare semplice. Inferenza sui parametri, diagnostica sui residui. Analisi su casi reali attraverso il pacchetto statistico R. Modello di regressione lineare multiplo. Inferenza sui parametri. Analisi dei residui. Il problema della Multicollinearità. La scelta delle variabili. Variabili esplicative di tipo qualitativo. Analisi su casi reali attraverso il pacchetto statistico R. Materiale didattico teorico: Il materiale didattico verrà fornito a lezione sottoforma di dispense.

 

Modalità d’esame:

L’esame consisterà in una prova scritta suddivisa in due parti, una parte teorica ed una parte pratica con il pacchetto statistico R.

Materiale didattico: 

xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

 

Research Interests

  • Quantile regression
  • Graphical models
  • Random Forest
  • Hidden Markov Models
  • Risk measures and models
  • Time Series analysis

Recent most important publications

  • (2024) Expectile hidden Markov regression models for analyzing cryptocurrency returns (with B. Foroni and L. Merlo) Statistics and Computing vol. 34, article 66
  • (2024) Inter-order relations between equivalence for Lp-quantiles of a Student t distribution (with V. Bignozzi and L. Merlo) forthcoming  Insurance Mathematics and Economics 
  • (2024) Quantile and Expectile copula-based hidden Markov regression models for the analysis of the cryptocurrency market (with B. Foroni and L. Merlo) under revision Statistical Modelling
  • (2023) Unifiend unconditional regression for multivariate quantiles, M-quantiles and expectiles (with L. Merlo, N. Salvati, N. Tzavidis) online Journal of the American Statistical Society
  • (2023) M-quantile regression shrinkage and selection via the Lasso and Elastic Net to asses the effect of meteorology and traffic on air quality (with F. Pantalone, G. Ranalli and N. Salvati)  Biometrical Journal  vol. 65nal
  • (2023) Mixed-frequency quantile regression to forecast Value at Risk and Expected Shortfall (with V. Candila and G. Gallo)  Annals of Operational Research
  • (2023) Neural Network for quantile claim ammount estimation: a quantile regression approach (with A. Laporta and S. Levantesi)  Annals of Actuarial Science 
  • (2023) A Neural network approach to price correleted health risks (with A. Laporta and S. Levantesi) submitted Annals of Operational Research
  • (2023) Estimating causal quantile exposure response functions via matching (with L. Merlo, F. Dominci, N. Salvati, X. Wu) submitted to Biometrika 
  • (2023) Quantile mixed graphical models with an application to mass public shooting in USA  (with L. Merlo, M. Geraci) submitted to Annals of Applied Statistics
  • (2023) Can the use of new Haemostatic agent improve short term outcomes of surgicl trated Non-small cell lung cancer patients? (Ricciardi, S., Cardillo, G et al.) Journal of Thoracic Oncology 18 S 511
  • (2022) Marginal M-quantile regression for multivariate dependent data (with L. Merlo, N. Salvati, N. Tzavidis) on line version, Computational Statistics and Data Analysis
  • (2022) Quantile Hidden Semi-Markov models for multivariate time series (with L.Merlo, A. Maruotti, A. Punzo) on line versione Statistics and Computing
  • (2022) The Network of Commodity Risk (with B. Foroni, G. Morelli) on line version Energy System
  • (2022) Sparse simulation-based estimation built on quantiles (with. P. Stolfi, M. Bernardi) on line version Econometrics and Statistics 
  • (2022) Quantile Mixed Hidden Markov Models for multivariate longitudinal data: an appllication to children's strenghts and difficulties questionannaire scores. (with L. Merlo and N. Tzavidis) Journal of the Royal Statistical Society Series C, 71, pp.417-448
  • (2022) Quantile Graphical Lasso: an application to cryptocurrencies, commodities and stock indexes (with B. Foroni and L. Merlo) under revision for Annals of Aplied Statistics 
  • (2021) COVID-19 after lung resection in Northern Italy (with M. Scarci, F. Raveglia, L. Merlo, G. Cardillo et al.) online version Seminars in Thoracic and Cardiovascular Surgery
  • (2021) Forecasting VaR and ES using a joint quantile regression and implications in portfolio allocation (with L. Merlo and V. Raponi) online version Journal of Banking and Finance
  • (2021) Unified Bayesian Conditional Autoregressive Risk Measures using the Skew Exponential Power Distribution (with M. Bottone and M. Bernardi) Statistical Methods and Applications, 30 pp.1079-1107
  • (2021) Two-part quantile regression models for semi continuous longitudinal data: a finite mixture approach (with L. Merlo and A. Maruotti) online version Statistical Modeling
  • (2021)  Hidden semi-Markov-switching quantile regression for time series (with A. Maruotti and L. Sposito)  online version Computational Statistics and Data Analysis
  • (2021) Option Pricing. Zero Lower Bound and COVID-19 (with G. Morelli) Risks 9 (9), 167
  • (2021) Nonthyroidal illness syndrome (NTIS) in severe COVID-19 patients: role of T3 on the Na/K pump gene expression and on hydroelectrolytic equilibrium (with S. Sciacchitano et al.)  Journal of Translation Medicine, 19 pp.1-18
  • (2021) Multivariate Analysis of Energy Commodieties during the COVID-19 Pandemic: Evidence for a Mixed Frequency Approach (with M Andreani, G. Morelli, V. Candila) Risks 9(8), 144
  • (2020) 3D Reconstruction Model of an Extra-Abdominal Desmoid Tumor: a Case Study (with F. Marinozzi, F. Carleo, S. Novelli, M. Di Martino, G. Cardillo, F. Bini) Frontiers in Bioengineering and Biotechnology vol.8 1-5
  • (2020) Sectorial Decomposition of CO2 world emission (with Luca Merlo and Valentina Raponi) International Review of Enviromental and Resource Economics pp. 197-238
  • (2020) Dynamica Model Averaging for Bayesian Quantile Regression (with M. Bernardi, R. Casarin, B. Millet) submitted to Annals of Operational Research arXiv:1602.00856
  • (2020) Large deviations for method of quantile estimators of one dimensional parameters, (with V. Bignozzi and C. Macci) Communications in Statistics- Theory and Methods pp.1132-1157
  • (2019) Joint estimation of conditional quantiles in multivariate linear regression models with an application to financial distress (with V. Raponi) Journal of Multivariate Analysis 173, 70-84

Postdoc and Phd students

  • Maria Saiz
  • Ismail Yanilmez
  • Beatrice Foroni
  • Martin Rossi
  • Valentina Raponi
  • Luca Merlo
  • Alessandro Laporta
  • Marco Bottone
  • Valeria Bignozzi
  • Mauro Bernardi