L.Zhu@lse.ac.uk
Working Papers
This paper proposes a novel approach to disentangling a Fed information effect from an exogenous monetary shock using high-frequency interest rate changes around a monetary announcement. The approach relies on the different ways these two factors change short-term interest rates. The key to identification is to consider monetary announcements and macroeconomic data releases together and let the latter inform us how interest rates respond to news on economic fundamentals. My measure of the information component of Fed announcements is strongly correlated with the difference between market forecasts and the Fed’s own forecasts. It has the advantage over measures based on Fed forecasts in that researchers have to wait five years for the release of the Fed forecasts, whereas the measure proposed here can be constructed in real-time from publicly available data. When one removes the information component from the response to monetary announcements, a pure policy shock has a bigger effect on the economy than suggested using a high-frequency policy instrument with no adjustments. A tightening monetary shock that raises the three-month-ahead fed funds futures rate by 1% leads industrial production to decline nearly 2.5% ten months after the shock. The CPI also responds to monetary policy more quickly than is implied by other estimates.
The US nominal Treasury yield curve has a particularly high volatility around certain macroeconomic data releases when initial and revised estimates of key macroeconomic indicators are announced. This means that economic agents do not know the true value of economic fundamentals until several months after the fact. I model macroeconomic variables in a New-Keynesian framework with fundamental shocks realized contemporaneously but only known at a future date. I model macroeconomic variables in a New-Keynesian framework with fundamentals shocks realized contemporaneously but only known at a future date. To respect potentially ill-behaved data revisions, I model initial and revised released numbers as noisy signals of true macroeconomic variables. I study bond market responses to data releases in light of these frictions in a no-arbitrage affine term structure model of interest rates with latent macroeconomic variables. The model identifies agents' expectations of aggregate economic outcomes after each data release and explains the heterogeneous magnitude of yield curve response to various types of releases. I find that the most important data releases seem to be nonfarm payrolls and the advance estimate of GDP. The importance of nonfarm payrolls comes primarily from its role in signaling the output growth. The release of advance estimates of GDP seems to convey substantial information about the equilibrium real rate of interest.
Work in Progress
I propose a state-space approach to decomposing a stock's idiosyncratic volatility (IV) into a common component and an idiosyncratic one. The IV of a stock is defined as the stochastic volatility of its idiosyncratic return. Unlike a simple average of IV's in the cross section or the first principal component of IV's in a panel in existing studies, the common component here is an AR(1) process and captures the persistence of IV's at the daily frequency. Using a panel of stocks in the Dow Jones Industrial Average Index from 1963 to 2009, I estimate the model with a Bayesian Gibbs sampling procedure. The model is found to fit the panel of IV's in sample better than a GARCH(1,1) model and a principal component approach. It also forecasts future levels of IVs better than a GARCH(1,1) model in the medium- to long-run. Instead of taking a static view of IV, I assess the pricing implications of the unpredictable part of the common component in the cross section of stock returns at the daily frequency. I find that the common component is not priced in my sample and sub-samples.
Teaching
FM437 Financial Econometrics, MT 2022 FM321 Risk Management and Modeling / FM320 Quantitative Finance, MT 2022