Essays on macroeconomic expectations and on economic policy uncertainty

• Ano de defesa: 2024
• Autor(a) principal: Géa, Cristiane
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Palavras-chave em Português: Política macroeconômica; Incerteza (Economia); Ativo financeiro; Mercado de capital; Previsão econômica; Events detection; Economic events; Macroeconomic policy; Uncertainty; ccapm model; Asset pricing; Stock market; Forecast; Macroeconomic forecast
• Link de acesso: https://app.uff.br/riuff/handle/1/35170

This thesis is a contribution to the literature on the evaluation of macroeconomic forecasts and the effects that economic policy uncertainty exerts on the stock market and on economic and financial time series. The main objectives are: (i) to offer a novel approach to evaluate biases in inflation and output growth forecasts, considering that the shocks that occur in a given month do not affect forecasts for the current year and the next in exactly the same way; (ii) to investigate the effects of economic policy uncertainty on the Brazilian stock market; (iii) to analyze the pricing of innovations in the Brazilian stock market during periods of economic policy uncertainty; and (iv) to analyze the association between economic policy uncertainty shocks and the types of anomalies that may show up in Brazilian economic and financial time series. The data used to tackle these issues comprises the period between 2000 and 2021. Overall, the results show that (i) the new modeling framework for forecast revisions makes the rejection of the null of no biases more frequent, especially when projections made many months before the end of the target year are considered; (ii) economic policy uncertainty correlates negatively with current excess equity returns and positively with future excess equity returns; furthermore, it is not significantly related to future dividend growth rates and expected changes in discount rates; (iii) the innovations brought about by economic uncertainty episodes improve the pricing of financial assets, making them a significant risk factor; (iv) the trend and the change point anomalies are the most prominent transformations that affect economic and financial time series in Brazil; and (v) the application of the ensemble methodology contributes to a more accurate event detection when compared to the adoption of individual methods.