Modelling and forecasting time series in the BigData Era: high-dimensional and high-frequency data

• Grant number: 22/09122-0
• Support Opportunities: Regular Research Grants
• Start date: November 01, 2022
• End date: April 30, 2025
• Field of knowledge: Applied Social Sciences - Economics - Quantitative Methods Applied to Economics
• Principal Investigator: Carlos Cesar Trucios Maza
• Grantee: Carlos Cesar Trucios Maza
• Host Institution: Instituto de Matemática, Estatística e Computação Científica (IMECC). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil

Abstract

Big Data is characterized by volume, velocity, variety, and veracity of the data used, as well as the value provided after an appropriate data analysis. In a context of time series, Big Data implies analysing a large number of time series jointly (high dimension) in fine granularities (high frequency), and can even use data from external sources, such as social networks, newspapers and specialized forums. This has generated a growing demand for appropriate models and methodologies for modelling and forecasting time series in different fields, such as economics, finance, energy, retail, urban traffic and cybersecurity. Therefore, it is necessary the constant development/extension of models and methodologies that allow to properly capture the dynamics of this data, object of this research project. (AU)

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

TRUCIOS, CARLOS; TAYLOR, JAMES W.. A comparison of methods for forecasting value at risk and expected shortfall of cryptocurrencies. JOURNAL OF FORECASTING, v. N/A, p. 19-pg., 2022-11-23. (22/09122-0)