Jose Augusto Fiorucci
Universidade de Brasília - UnB, Statistic, Faculty Member
Accurate and robust forecasting methods for univariate time series are critical as the historical data can be used in the strategic planning of such future operations as buying and selling to ensure product inventory and meet market... more
Accurate and robust forecasting methods for univariate time series are critical as the historical data can be used in the strategic planning of such future operations as buying and selling to ensure product inventory and meet market demands. In this context, several competitions for time series forecasting have been organized, with the M3-Competition as the largest. As the winner of M3-Competition, the Theta method has attracted attention from researchers for its predictive performance and simplicity. The Theta method is a combination of other methods, which proposes the decomposition of the deseasonalized time series into two other time series called "theta lines". The first completely removes the curvatures of the data, thus accurately estimating the long-term trend. The second doubles the curvatures to better approximate short-term behavior. Several issues have been raised about the Theta method, even by its originators. They include the number of theta lines, their par...
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Abstract In this article, we proposed a new three parameter lifetime distribution motivated mainly by lifetime issues, which generalizes the Exponential Poisson distribution proposed by Cancho et al. (2011). We derive various standard... more
Abstract In this article, we proposed a new three parameter lifetime distribution motivated mainly by lifetime issues, which generalizes the Exponential Poisson distribution proposed by Cancho et al. (2011). We derive various standard mathematical properties of the proposed model including a formal proof of its probability density function and hazard rate function. The inference via the maximum likelihood approach is discussed. The performance of the maximum likelihood estimators, the likelihood ratio test and its power are studied by simulation. Finally, the proposed model is fitted to two real data sets and it is compared with several models.
Research Interests:
Research Interests:
Research Interests:
Investimentos baseados na análise técnica vêm sendo utilizados com maior frequência para examinar o desempenho estratégico das negociações automatizadas por meio de um robô (algoritmo) de investimentos, em particular, usando os... more
Investimentos baseados na análise técnica vêm sendo utilizados com maior frequência para examinar o desempenho estratégico das negociações automatizadas por meio de um robô (algoritmo) de investimentos, em particular, usando os indicadores Parabolic SAR e Fibonacci. Este trabalho utilizou a avaliação de cenários para posteriormente comparar seus resultados em relação à estratégia de buy and hold. Os cenários se diferenciam em relação à utilização de fatores de risco, timeframes e níveis de preço. Backtests foram realizados por um período compreendido entre Janeiro de 20015 e Abril de 2017 para comparar as estratégias. Como resultado, foi percebido que a utilização da análise técnica por meio da negociação automatizada pode resultar em lucros superiores à buy and hold. Entretanto, tal forma de negociação apresenta um alto nível de volatilidade.
Research Interests:
Research Interests:
Research Interests:
Research Interests:
Multivariate GARCH models are important tools to describe the dynamics of multivariate times series of financial returns. Nevertheless, these models have been much less used in practice due to the lack of reliable software. This paper... more
Multivariate GARCH models are important tools to describe the dynamics of multivariate times series of financial returns. Nevertheless, these models have been much less used in practice due to the lack of reliable software. This paper describes the {\tt R} package {\bf BayesDccGarch} which was developed to implement recently proposed inference procedures to estimate and compare multivariate GARCH models allowing for asymmetric and heavy tailed distributions.
ABSTRACT The main goal in this paper is to develop and apply stochastic simulation techniques for GARCH models with multivariate skewed distributions using the Bayesian approach. Both parameter estimation and model comparison are not... more
ABSTRACT The main goal in this paper is to develop and apply stochastic simulation techniques for GARCH models with multivariate skewed distributions using the Bayesian approach. Both parameter estimation and model comparison are not trivial tasks and several approximate and computationally intensive methods (Markov chain Monte Carlo) will be used to this end. We consider a flexible class of multivariate distributions which can model both skewness and heavy tails. Also, we do not fix tail behaviour when dealing with fat tail distributions but leave it subject to inference.
Research Interests:
Abstract Combination methods have performed well in time series forecast competitions. This study proposes a simple but general methodology for combining time series forecast methods. Weights are calculated using a cross-validation scheme... more
Abstract Combination methods have performed well in time series forecast competitions. This study proposes a simple but general methodology for combining time series forecast methods. Weights are calculated using a cross-validation scheme that assigns greater weights to methods with more accurate in-sample predictions. The methodology was used to combine forecasts from the Theta, exponential smoothing, and ARIMA models, and placed fifth in the M4 Competition for both point and interval forecasting.
Research Interests:
A growing field is related to automatized Time Series analysis, through complicated due to the dependence of observed and hidden dimensions often presented in these data types. In this report the problem is motivated by a Brazilian... more
A growing field is related to automatized Time Series analysis, through complicated due to the dependence of observed and hidden dimensions often presented in these data types. In this report the problem is motivated by a Brazilian financial company interested in unraveling relation structure explanation of the Japanese' CPI ex-fresh Food \& Energy across 157 economical exogenous variables, with very limiting data. The problem becomes more complex when considering that each variable can enter the model with lags of 0 to 8 periods, as well as an additional restriction of admitting only a positive relationship. This report discusses three possible treatments involving models for structured time series, the most relevant approach found in this study is a Dynamic Regression Model combined with a Stepwise algorithm, which allows the most relevant variables, as well as their respective lags, to be found and inserted in the model with low computational cost.