Aplicação de diferentes metodologias do Value at Risk (VaR) para mensuração do risco de mercado de uma carteira teórica de ações

• Autor(a) principal: Fernandes Júnior, Marcelo Aparecido
• Data de Publicação: 2022
• Tipo de documento: Trabalho de conclusão de curso
• Idioma: por
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://hdl.handle.net/11449/217400
• topic: Value at risk; Market risk; Stock portfolio; Historical simulation; Exponentially weighted moving average; GARCH; Extreme value theory; Valor em risco; Risco de mercado; Carteira de ações; Simulação histórica; Suavização exponencial; Volatilidade condicional; Teoria dos valores extremos

Value at Risk is a risk measure widely used by financial institutions to assist in the allocation of the minimum capital requirement. This study presents the regulation of risk management for financial institutions and the following VaR estimation methods: Historical Simulation – HS, Exponentially Weighted Moving Average – EWMA, Autoregressive Conditional Heterocedasticity – GARCH (GARCH Normal and Skew-tGARCH) and Extreme Value Theory – EVT. For the analysis, a theoretical portfolio was built containing the shares of Itaú, Bradesco, Petrobras and Vale, from which the time series of a period referring to the COVID-19 pandemic (january 2020 to september 2021) was observed, which served to predict the maximum potential losses with 99% confidence for the month of august and semptember 2021, according to each of the methods. The models were approved by Backtesting according to Kupiec's criterion, with the exception of two univariate EWMA models that were disapproved and the Theory of Extreme Values, which was not implemented due to its restrictions. The models showed satisfactory results even in a period of high volatility.