財務金融研究

Abstract

本論文包含三個部份。 第一部分是一個建構在參數不確定性下的資產配置模型。 在期望資產報酬率不確定的假設之下,我們發現對期望資產報酬率的估計誤差所引起的估計風險(Estimation Risk)可以部份的解釋國內資產偏誤(Home Bias Puzzle)。我們同時發現投資者的風險趨避程度愈高, 資產偏誤的程度會愈強烈。 第二部份主要是在研究預期通貨膨脹率的不確定性對資產配置與消費行為的影響。 我們發現投資者的實質消費行會因名目物價的改變而受到影響。同時我們也指出了對預期通貨膨脹率的估計風險將會減損投資者的實質消費進而減損消費者福利。在第三部份, 我們試圖建立一個由技術研發所引導的廠商內生成長模型。我們發現在特定的參數條件之下, 廠商的最適成長率會有負的規模效果, 亦即小廠的成長率會高於大廠。 在同樣的條件下我們亦解釋了為何成長率高的廠商會有較低的股票報酬。

This thesis is composed of three parts. In Chapter Two, I present an asset allocation model with unknown return on risky assets. I conclude that the effect of leaning about exact asset returns would make long-term investor reduce his holding on risky asset. Besides, with heterogeneous prior believes in estimated returns, the optimal portfolio weight would bias toward assets with higher prior confidence. This partly explains the home bias puzzle. In Chapter Three, I solve for an asset allocation problem under unobservable inflation rate. I show that the investor's optimal portfolio would have an additional hedging demand against the estimation risk of inflation. The estimation risk also reduces the investor's optimal consumption. It's worth to note that the change of nominal price level would affect the real consumption through the estimated inflation. In Chapter Four, I establish a production-based partial equilibrium model of equity price and equity return with technology progress. Under specific condition, I successfully show that equity return is negatively associated to technology growth rate. I also indicate that when R&D of new technology improves the profitability less effectively, the equity return would be relatively lower.