Browse DORAS
Browse Theses
Search
Latest Additions
Creative Commons License
Except where otherwise noted, content on this site is licensed for use under a:

Empirical investigation of nonlinear asset pricing kernel with human capital and housing wealth

Wang, Qing Mei (2011) Empirical investigation of nonlinear asset pricing kernel with human capital and housing wealth. Master of Business Studies thesis, Dublin City University.

Full text available as:

[img]
Preview
PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
2518Kb

Abstract

In a traditional framework, asset returns are captured by simple linear asset pricing models. They include Capital Asset Pricing Model (CAPM) and Fama-French threefactor model. However, the empirical study shows that the asset returns are fat tailed, that cannot be accurately predicted by normal distribution. Kurtosis and skewness should be considered when pricing those non-normal assets. Various literatures can be found focused on this topic. Bansal and Viswanathan (1993) and Chapman (1997) developed nonparametric model. They find that the nonparametric models perform better in explaining expected returns. Most recently, nonlinear asset pricing models developed by Dittmar (2002) shows more significantly improvements in return estimation, compared to the linear single and linear multi-factor models. In this study, I focus on an asset-pricing model of higher order risk factors and use polynomial pricing kernel to generate the empirical performance of a nonlinear model. This is an extension to both Bansal and Dittmar’s work, by extending the definition of the total wealth including human capital and housing wealth. This research work is novel and especially important to understand asset price behavior after year 2007, the credit crisis. Housing price growth rate is a very critical indicator for long-term investment, reflecting consumer confidence on the long-term global economy. It can be used to estimate the turning point for the recent economic down turn. In addition, since the credit crisis 2008 is triggered by liquidity shortage in banking systems, the level of housing price has direct impact on the balance sheet of those banking sectors. The higher the house price, the more willingness banks have to release the credit to the market. The housing wealth factor can be used to estimate when the credit crunch will disappear and global economy gets fully recovered. In this study, the risk factors that represent the aggregate wealth in the economic are tested. The best possible proxy of return on the total wealth is discussed. The thesis can be divided into 2 parts. In the first part of my thesis, a higher order moment model to explain the asset price behaviour is developed. Similar to the work presented by Dittmar (2002), pricing kernel is approximated using Taylor Series expansion and Hansen-Jagannathan (1997) weighting matrix. The time-varying coefficients with respected sign of coefficients are estimated. Housing factor is added to extend the model, as we believe that housing plays an important role in the return on aggregate wealth. In the second part of my thesis, I test models in three time periods. They include Dittmar’s period from 1963 to 1995, the full sample period from 1963 to 2009 and recent period from 1996 to 2009. Our results confirm that nonlinear models outperform than linear models in explaining the cross section of returns. The higher order risk factors give the magnitude improvement in model fitting. This is consistent with the result given by Dittmar (2002). Moreover, my results conclude that the models with the housing wealth included performs significantly better than the models with human capital only.

Item Type:Thesis (Master of Business Studies)
Date of Award:November 2011
Refereed:No
Supervisor(s):Gallagher, Liam
Uncontrolled Keywords:Nonlinear asset pricing kernel; human capital; housing wealth
Subjects:Business > Finance
DCU Faculties and Centres:DCU Faculties and Schools > DCU Business School
Use License:This item is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 3.0 License. View License
ID Code:16512
Deposited On:28 Nov 2011 11:17 by Rachel Keegan. Last Modified 28 Nov 2011 11:52

Download statistics

Archive Staff Only: edit this record