Count time series with excess zeros: A Bayesian perspective using zero-adjusted distributions

Models for count data which are temporally correlated have been studied using many conditional distributions, such as the Poisson distribution, and the insertion of different dependence structures. Nonetheless, excess of zeros and over dispersion may be observed during the counting process and need to be considered when modelling and choosing a conditional distribution. In this paper, we propose models for counting time series using zero-adjusted distributions by inserting a dependence structure following the ARMA(p, q) process on a Bayesian framework. We perform a simulation study using the proposed Bayesian analysis and analyse the monthly time series of the number of deaths due to dengue haemorrhagic fever (ICD-A91) in Brazil.

Volatility of intraday financial data: Multiscale Ibovespa behavior under to the COVID-19 pandemic

In financial markets, volatility modeling has been a strategy widely used because it reflects uncertainties about changes in asset prices. Incorporating peculiarities of financial series, this study estimated the volatility for the intraday index of the Brazilian stock market (Ibovespa) using ARIMA-APARCH models in different time frequencies with the aid of the wavelet MODWT decomposition technique. This work proposes an analysis of the impacts of the frequency components on the behavior of the volatility of intraday returns using the series of details wavelet in different time horizons, in an atypical period in the global financial markets, generated by the COVID-19 pandemic. The empirical results suggest low unconditional volatility and strong signs of persistence in all analyzed frequencies. The asymmetry in volatility is evidenced in the higher frequencies, the leverage effect being present only in the series of details with variations of 15-120 min., which is corroborated with the results obtained with the reconstructed series. The evidenced behaviors have an impact on the elaboration of short-term investment strategies and risk management, since the positive and negative shocks, such as those given by the world pandemic of COVID-19, have different impacts on the volatility of returns in shorter periods. The information obtained can contribute to the analysis of future atypical events in the Brazilian stock market, supporting the decision-making of economic agents.

Asymmetric dependence of intraday frequency components in the Brazilian stock market

The multivariate dependence plays an important role in financial instrument management. Due to the inherent characteristics in the financial market, such as heavy tails in the returns unconditional distribution and asymmetry between gain and loss, we obtained the asymmetric dependence structure in different short-term variation scales based on the wavelet technique MODWT. The study sought to capture the relations between financial returns represented by its frequency components. Intraday returns series was used in the 15-min sampling interval from stocks and applied the D-Vine pair-copula to decompose in trade frequencies of 15 min, 1 h, 1 day, and 1 week with margin adjustments of ARIMA-APARCH class and BB7 copula function, responsible for measuring the dependence on tails. The results indicated the prevalence of a high dependence during market upturns, rising over the analyzed frequencies. Being an important tool in financial management and allowing short-term strategies of diversification.

Ocorrência de sinistros em veículos no sul de Minas Gerais: uma abordagem via modelos de classificação

Com as mudanças nos padrões de risco, novos produtos de seguros são disponibilizados no mercado, atendendo as demandas do consumidor. Consequentemente, os modelos de precificação são reestruturados de modo a gerenciar os níveis de risco e estabelecer prêmios que mantenham o bem estar atuarial, alocando apólices em carteiras através de modelos de classificação e clusterização. Este trabalho analisou o desempenho dos modelos logístico e Random forests na classificação de ocorrências de sinistros do tipo colisão por perda total no sul de Minas Gerais utilizando amostras de treino originais e artificiais via método de reamostragem ROSE. Considera-se a perda total de um veículo quando os custos de reparos do sinistro de um mesmo evento superarem um percentual estabelecido contratualmente. Como resultado, obteve-se que os modelos com amostra artificial apresentaram resultados de acurácia superiores aos demais, indicando a melhoria nas taxas de acurácia através de métodos de reamostragem durante o treino.