Apergis, Iraklis (2022) Essays on the Predictive Content of Option Prices and Tail Uncertainty of Asset Returns. Doctor of Philosophy (PhD) thesis, University of Kent,. (doi:10.22024/UniKent/01.02.97261) (KAR id:97261)
PDF
Language: English
This work is licensed under a Creative Commons Attribution 4.0 International License.
|
|
Download this file (PDF/0B) |
Preview |
Official URL: https://doi.org/10.22024/UniKent/01.02.97261 |
Abstract
This thesis explores two key elements that have been the subject of academic and practical review for many years. First is the identification of predictors of the equity premium, and second is the construction of a profitable portfolio. The first two empirical studies employ established techniques that can categorise predictors as either strong, relatively strong, or weak, based on their predictive performance across different parts of their distribution (quantiles), alongside other techniques that combine forecasts from different univariate, or multivariate models. The academic literature for the last years has conformed around the validity of certain economic indicators, which this thesis aims to expand by employing information from the option markets. In the first exercise, the information from the CBOE indices is targeted, with a somewhat weak performance in generating out-of-sample point forecasts. The VIX index was the dominantly selected variable out of all the CBOE indices. While there were no significant values across allquantiles consistently, all in all, there is evidence that some of the indices have predictive information on a few selected quantile levels. The second exercise further expands from the indices and employs directly option-implied information from the market. These,now daily frequented, variables were consistently selected by the penalising out-of-sample algorithm, and proved to be selected over any of the other economic variables that the literature had already established. Density forecasts were as well created. However, there was some evidence to suggest that only a minority of the option-implied information could provide significant density forecasts. The final exercise inspired by the density forecasts employs a Bayesian approach in order to estimate future values and risk of the asset's distributional parameters. This application allowed for creating a constantly short-long position that proved to yield a positive cumulative return by the end of the trading positions.
Item Type: | Thesis (Doctor of Philosophy (PhD)) |
---|---|
Thesis advisor: | Voukelatos, Nikolaos |
Thesis advisor: | Oberoi, Jaideep |
DOI/Identification number: | 10.22024/UniKent/01.02.97261 |
Uncontrolled keywords: | options forecasting equity premium tail uncertainty |
Subjects: | H Social Sciences |
Divisions: | Divisions > Kent Business School - Division > Department of Accounting and Finance |
SWORD Depositor: | System Moodle |
Depositing User: | System Moodle |
Date Deposited: | 04 Oct 2022 07:40 UTC |
Last Modified: | 05 Oct 2022 11:17 UTC |
Resource URI: | https://kar.kent.ac.uk/id/eprint/97261 (The current URI for this page, for reference purposes) |
- Link to SensusAccess
- Export to:
- RefWorks
- EPrints3 XML
- BibTeX
- CSV
- Depositors only (login required):