This dissertation studies the relationship between the term structure of interest rates, monetary policy, and macroeconomy. The first chapter, A Parsimonious No- Arbitrage Term Structure Model that is Useful for Forecasting, offers a solution to a well-known puzzle in the term structure literature. The puzzle is that while the level, slope and curvature (or the first three principal components of yields) can quite accurately summarize the cross-section of yields at any point in time, different functions of interest rates and other macroeconomic variables appear to be helpful when the goal is to predict future interest rates. My paper proposes a parsimonious representation to capture this feature in a large dataset. In the first step, I run reduced rank regressions of one-year excess returns on a panel of 131 macroeconomic variables and initial forward rates from 1964 to 2007. I find that a single linear combination of macroeconomic variables and forward rates can predict excess returns on two- to five-year maturity bonds with R- squared up to 0.71. The forecasting factor subsumes the tent-shaped linear combination of forward rates constructed by Cochrane and Piazzesi (2003) and explains excess returns better. In the second step, I estimate a restricted Gaussian Affine Term Structure Model (GATSM) with the level, slope and curvature commonly used by most term structure models along with the forecasting factor. Restrictions are derived based on the fact that while cross-sectional information in yields is spanned by the level, slope and curvature, cross-sectional information in expected excess returns is spanned by the forecasting factor. Compared with a conventional GATSM only including the level, slope and curvature, the restricted four-factor GATSM generates plausible countercyclical term premia. The second and third chapter focus on the recent zero lower bound (ZLB) period. In the second chapter, Measuring the Macroeconomic Impact of Monetary Policy at the Zero Lower Bound, coauthored with Cynthia Wu, we employ an approximation that makes a nonlinear shadow rate term structure model (SRTSM) extremely tractable for analysis of an economy operating near the zero lower bound for interest rates. We show that such a model offers a better description of the data compared to the widely used GATSM. Moreover, the model can be used to summarize the macroeconomic effects of unconventional monetary policy at the ZLB. Using a simple factor-augmented vector autoregression (FAVAR), we show that the shadow rate calculated by our model exhibits similar dynamic correlations with macro variables of interest in the period since 2009 as the fed funds rate did in data prior to the Great Recession. This result gives us a tool for measuring the effects of monetary policy under the ZLB, using either historical estimates based on the fed funds rate or less precisely measured estimates inferred solely from the new data for the shadow rate alone. We show that the Fed has used unconventional policy measures to successfully lower the shadow rate. Our estimates imply that the Fed's efforts to stimulate the economy since 2009 have succeeded in lowering the unemployment rate by 0.13% relative to where it would have been in the absence of these measure. The third chapter, Effects of Unconventional Monetary Policies on the Term Structure of Interest Rates, offers a complete characterization of effects of unconventional monetary policies on interest rates by examining policies' impacts on the whole yield curve. I make use of the SRTSM to summarize all interest rates with factors of lower dimension so that I can capture responses of all interest rates in a parsimonious way. By investigating how policy announcements affect the three factors and then the whole forward curve accordingly, I find that during the ZLB period, forward rate with short maturities are constrained, while forward rates with long maturities still respond to policy announcements. Following each easing (tightening) policy announcement, long forward rates would decrease (increase) by 10 basis points on average
