Essays on Asset Pricing and Empirical Estimation PDF Download

Author: Pooya Nazeran
Publisher:
ISBN:
Category :
Languages : en
Pages : 138

Get Book Here

Book Description
Abstract: A considerable portion of the asset pricing literature considers the demand schedule for asset prices to be perfectly elastic (flat). As argued, asset prices are determined using information about future payoff distribution, as well as the discount rate; consequently, an asset would be priced independent of its available supply. Furthermore, such a flat demand curve is considered to be a consequence of the Efficient Market Hypothesis. My dissertation evaluates and questions the factuality of these assertions. I approach this problem from both an empirical and a theoretical perspective. The general argument is that asset prices do respond to supply-shocks; and changes in aggregate demand, stemming from preference changes, new international investments, or quantitative easing by the Fed, can result in price changes. Hence, asset prices are determined by both demand and supply factors. In the first essay, "Downward Sloping Asset Demand: Evidence from the Treasury Bills Market," I report on my empirical study which establishes the existence of a downward sloping demand curve (DSDC) in the T-bill market. In the second essay, "Asset Pricing: Inelastic Supply," I examine the theoretical issues concerning a downward sloping demand curve. I begin by clarifying a common confusion in the literature, namely, that many asset pricing models imply a flat demand curve. I show that the prominent asset pricing models, including Capital Asset Pricing Model (CAPM), Arbitrage Pricing Theory (APT) and Consumption Capital Asset Pricing Model (CCAPM), all have an underlying DSDC. I further show that, while these models imply the relevance of supply, they are inconvenient as a vehicle for the estimation and analysis of the DSDC in the data. For those purposes, I develop an asset pricing framework based on the stochastic discount factor framework, specifically designed with a DSDC at its heart. I end the essay with a discussion of the framework's implications and applications. In the third essay I develop on the Factor-Augmented Vector-Autoregression (FAVAR) literature, proposing a bias-corrected method. As implemented in the literature, the Principal Component Analysis stage of FAVAR introduces a classical-error-in-variable problem which leads to bias. I propose an instrument-based method for bias correction.