O INSPER E ESTE REPOSITÓRIO NÃO DETÊM OS DIREITOS DE USO E REPRODUÇÃO DOS CONTEÚDOS AQUI REGISTRADOS. É RESPONSABILIDADE DO USUÁRIO VERIFICAR OS USOS PERMITIDOS NA FONTE ORIGINAL, RESPEITANDO-SE OS DIREITOS DE AUTOR OU EDITORDewachter, HansIania, LeonardoLyrio, Marco Túlio Pereira2023-07-172023-07-172011https://repositorio.insper.edu.br/handle/11224/5837This paper uses an affine term structure model that incorporates macroeconomic and financial factors to study the term premium in the U.S. bond market. The results corroborate the known rejection of the expectation hypothesis and indicate that one factor, closely related to the Cochrane and Piazzesi (2005) factor (the CP factor), is responsible for most of then variation in bond premia. Furthermore, the model-implied bond premia are able to explain around 32% and 40% of the variability of one- and two-year excess returns and their out-of-sample performance is comparable to the one obtained with the CP factor. The model is also used to decompose yield spreads into an expectations and a term premium componente in order to forecast GDP growth and ináation. Although this decomposition does not seem important to forecast GDP growth it is crucial to forecast ináation for most forecasting horizons. Also, the inclusion of control variables such as the short-term interest rate and lagged variables does not drive out the predictive power of the yield spread decomposition.44 p.DigitalInglêsworking paperMacro-Önance modelTerm structureForecastingBEWP 132/2011