Term Structure Dynamics, Macro-Finance Factors and Model Uncertainty

This paper extends the Nelson-Siegel linear factor model by developing a flexible macro-finance framework for modeling and forecasting the term structure of US interest rates. Our approach is robust to parameter uncertainty and structural change, as we consider instabilities in parameters and volatilities, and our model averaging method allows for investors' model uncertainty over time. Our time-varying parameter Nelson-Siegel Dynamic Model Averaging (NS-DMA) predicts yields better than standard benchmarks and successfully captures plausible time-varying term premia in real time. The proposed model has significant in-sample and out-of-sample predictability for excess bond returns, and the predictability is of economic value.

Stock returns, term structure, inflation and real activity: An international perspective

This paper analyses the empirical interdependences among assetreturns, real activity and inflation from a multicountry and internationalpoint of view. We find that nominal stock returns are significantly relatedto inflation only in the US, that the US term structure of interest ratespredicts both domestic and foreign inflation rates while foreign termstructures do not have this predictive power and that innovations in inflationand exchange rates induce insignificant responses of real and financialvariables. An interpretation of the dynamics and some policy implicationsof the results are provided.

Machine learning analysis and modeling of interest rate curves

The present research deals with the review of the analysis and modeling of Swiss franc interest rate curves (IRC) by using unsupervised (SOM, Gaussian Mixtures) and supervised machine (MLP) learning algorithms. IRC are considered as objects embedded into different feature spaces: maturities; maturity-date, parameters of Nelson-Siegel model (NSM). Analysis of NSM parameters and their temporal and clustering structures helps to understand the relevance of model and its potential use for the forecasting. Mapping of IRC in a maturity-date feature space is presented and analyzed for the visualization and forecasting purposes.

Relative influence of age, resting heart rate and sedentary life style in short-term analysis of heart rate variability

In order to assess the relative influence of age, resting heart rate (HR) and sedentary life style, heart rate variability (HRV) was studied in two different groups. The young group (YG) consisted of 9 sedentary subjects aged 15 to 20 years (YG-S) and of 9 nonsedentary volunteers (YG-NS) also aged 15 to 20. The elderly sedentary group (ESG) consisted of 16 sedentary subjects aged 39 to 82 years. HRV was assessed using a short-term procedure (5 min). R-R variability was calculated in the time-domain by means of the root mean square successive differences. Frequency-domain HRV was evaluated by power spectrum analysis considering high frequency and low frequency bands. In the YG the effort tolerance was ranked in a bicycle stress test. HR was similar for both groups while ESG showed a reduced HRV compared with YG. Within each group, HRV displayed a negative correlation with HR. Although YG-NS had better effort tolerance than YG-S, their HR and HRV were not significantly different. We conclude that HRV is reduced with increasing HR or age, regardless of life style. The results obtained in our short-term study agree with others of longer duration by showing that age and HR are the main determinants of HRV. Our results do not support the idea that changes in HRV are related to regular physical activity.

Moderate-term reproducibility of heart rate variability during rest and light to moderate exercise in children

Previous studies have demonstrated the high reproducibility of heart rate variability (HRV) measures in adults while little information exists concerning HRV reproducibility in children. Subsequently, the aim of the current study was to examine the moderate-term reproducibility of heart rate and frequency domain measures of HRV during rest and light to moderate exercise in children. Ten healthy children (6 males, 4 females) aged between 7 and 12 years of age volunteered for this study with HRV recordings obtained during supine rest and three treadmill walking exercise work rates (≤60% maximum heart rate), initially and then 8 weeks later. Differences (P < 0.05) between variables were examined using paired t-tests or Wilcoxon signed rank tests while reliability and reproducibility were examined by intraclass correlation coefficients (ICC), coefficients of variation (CV), and mean bias ratio and ratio limits of agreement (LOA). Heart rate and all measures of HRV at rest and exercise were unchanged after 8 weeks. Significant ICC were documented primarily during rest (0.72-0.85) while weaker relationships (-0.02-0.87) were evident during exercise. A large range of CV was identified during rest (6-33%) and exercise (3-128%) while the ratio LOA were variable and substantial (1.04-2.73). Despite similar HRV over an 8-week period, variable ICC and sizable CV and ratio LOA indicate moderate to poor reproducibility of HRV in children, particularly during light to moderate exercise. Studies examining HRV in children should include age- or maturation stage-matched control participants to address the age-related change in HRV and inadequate HRV reliability.

Testing the long-run implications of the expectation hypothesis using cointegration techniques with structural change

This paper investigates whether or not multivariate cointegrated process with structural change can describe the Brazilian term structure of interest rate data from 1995 to 2006. In this work the break point and the number of cointegrated vector are assumed to be known. The estimated model has four regimes. Only three of them are statistically different. The first starts at the beginning of the sample and goes until September of 1997. The second starts at October of 1997 until December of 1998. The third starts at January of 1999 and goes until the end of the sample. It is used monthly data. Models that allows for some similarities across the regimes are also estimated and tested. The models are estimated using the Generalized Reduced-Rank Regressions developed by Hansen (2003). All imposed restrictions can be tested using likelihood ratio test with standard asymptotic 1 qui-squared distribution. The results of the paper show evidence in favor of the long run implications of the expectation hypothesis for Brazil.

Essays in macroeconomy and dynamic term-structure models

This thesis is composed of three articles with the subjects of macroeconomics and - nance. Each article corresponds to a chapter and is done in paper format. In the rst article, which was done with Axel Simonsen, we model and estimate a small open economy for the Canadian economy in a two country General Equilibrium (DSGE) framework. We show that it is important to account for the correlation between Domestic and Foreign shocks and for the Incomplete Pass-Through. In the second chapter-paper, which was done with Hedibert Freitas Lopes, we estimate a Regime-switching Macro-Finance model for the term-structure of interest rates to study the US post-World War II (WWII) joint behavior of macro-variables and the yield-curve. We show that our model tracks well the US NBER cycles, the addition of changes of regime are important to explain the Expectation Theory of the term structure, and macro-variables have increasing importance in recessions to explain the variability of the yield curve. We also present a novel sequential Monte-Carlo algorithm to learn about the parameters and the latent states of the Economy. In the third chapter, I present a Gaussian A ne Term Structure Model (ATSM) with latent jumps in order to address two questions: (1) what are the implications of incorporating jumps in an ATSM for Asian option pricing, in the particular case of the Brazilian DI Index (IDI) option, and (2) how jumps and options a ect the bond risk-premia dynamics. I show that jump risk-premia is negative in a scenario of decreasing interest rates (my sample period) and is important to explain the level of yields, and that gaussian models without jumps and with constant intensity jumps are good to price Asian options.

Hedging options in a garch environment: testing the term structure of stochastic volatility models

This paper develops a methodology for testing the term structure of volatility forecasts derived from stochastic volatility models, and implements it to analyze models of S&P500 index volatility. U sing measurements of the ability of volatility models to hedge and value term structure dependent option positions, we fmd that hedging tests support the Black-Scholes delta and gamma hedges, but not the simple vega hedge when there is no model of the term structure of volatility. With various models, it is difficult to improve on a simple gamma hedge assuming constant volatility. Ofthe volatility models, the GARCH components estimate of term structure is preferred. Valuation tests indicate that all the models contain term structure information not incorporated in market prices.

Bahamas and Barbados: empirical evidence of interest rate pass-through

Includes bibliography

Forecasting the term structure of volatility of crude oil price changes

Peer reviewed

Sinalização de política monetária e movimentos na estrutura a termo da taxa de juros no Brasil

O objetivo deste artigo é examinar como as decisões de taxa de juros básica no Brasil (um forte mecanismo de sinalização em política monetária) afetam a estrutura a termo da curva de juros. Diferentemente de outros trabalhos sobre o caso brasileiro, este avalia a evolução da previsibilidade das decisões de política monetária após a introdução do regime de metas de inflação e, também, compara esta evolução com outros países. A metodologia utilizada é um estudo de eventos em 2 períodos distintos: entre jan/2000 e ago/2003, após a introdução do regime de metas de inflação, e entre set/2003 e jul/2008, quando o regime de metas atinge certa maturidade. Os resultados indicam que: 1) os efeitos surpresa na curva de juros estão menores; 2) o poder explicativo das ações de política monetária aumentou; 3) o mercado tem efetuado o ajuste das expectativas de decisão sobre a taxa de juros com antecedência de 3 dias; 4) a previsibilidade e transparência das decisões de política monetária no Brasil aumentaram e estão próximas daquelas observadas nos EUA e Alemanha e superiores ao caso italiano e britânico.

A estrutura a termo da taxa de juros no Brasil: testando relações de codependência de ordem zero

Este trabalho testa a existência de relações de codependência de ordem zero em spreads formados a partir da estrutura a termo da taxa de juros no Brasil. O objetivo é verificar se existem combinações lineares dos spreads que geram um processo ruído branco contemporâneo. Essas combinações lineares poderiam ser utilizadas para a previsão de taxas de juros futuras dado que desvios destas relações estáveis implicariam em um movimento futuro das taxas de juros no sentido de restabelecer o equilíbrio. O modelo de Nelson e Siegel (1987) serve de base teórica para os testes empíricos. Ao verificar a hipótese de codependência de ordem zero é possível também analisar premissas quanto aos parâmetros do modelo em relação à estrutura a termo da taxa de juros no Brasil. As evidências obtidas a partir dos resultados empíricos apontam na rejeição da hipótese de codependência de ordem zero e, consequentemente, na impossibilidade de definir as combinações lineares mencionadas. Esta constatação pode estar relacionada aos períodos de instabilidade presentes na amostra ou na existência de codependência de ordem superior a zero.

Relação entre as componentes principais da estrutura a termo da taxa de juros brasileira e as variáveis macroeconômicas

Este trabalho observa como as variáveis macroeconômicas (expectativa de inflação, juro real, hiato do produto e a variação cambial) influenciam a dinâmica da Estrutura a Termo da Taxa de Juros (ETTJ). Esta dinâmica foi verificada introduzindo a teoria de Análise de Componentes Principais (ACP) para capturar o efeito das componentes mais relevantes na ETTJ (nível, inclinação e curvatura). Utilizando-se as estimativas por mínimos quadrados ordinários e pelo método generalizado dos momentos, foi verificado que existe uma relação estatisticamente significante entre as variáveis macroeconômicas e as componentes principais da ETTJ.