Back
Texte intégral
Doctoral thesis

Essays in asset pricing

    26.05.2014

148 p

Thèse de doctorat: Università della Svizzera italiana, 2014

Heterogeneous beliefs
Systemic disasters
Lucas orchard
Variance risk premium
Correlation risk premium
Predictability
Predictive regression
Present-value model
State-space model
Bootstrap
Likelihood ratio test
English My dissertation consists of three main chapters and focuses on two recent strands of research in asset pricing, namely heterogeneous beliefs about rare event risk and present-value models for predictability of market returns and dividend growth. The first chapter studies the asset pricing implications of investor disagreement about the probability of a systemic disaster. I start from a structural economy with multiple assets and heterogeneous beliefs on systemic rare event risk, in order to understand how fear and risk sharing mechanisms affect excess returns on equity and pure variance positions and the relation between them, both at an aggregate level and in the cross section of stock returns. The second chapter proposes a bootstrap methodology to test predictability hypotheses in the context of present-value models with latent expectation processes for returns and dividends. We show that the test is asymptotically valid and has good finite-sample properties while conventional tests strongly over-reject the null of no predictability in small samples, due to restrictive distributional assumptions. The third chapter proposes a new class of tractable present-value models with latent dividend and return processes and time-varying cash flow and discount rate risks, to study the joint predictability features of dividends, returns and their second moments.
Language
  • English
Classification
Economics
License
License undefined
Identifiers
Persistent URL
https://n2t.net/ark:/12658/srd1318592
Statistics
Document views: 139 File downloads:
  • 2014ECO005.pdf: 102