Use this URL to cite or link to this record in EThOS:
Title: Essays on predictability & excess profitability of quantitative methods : modelling implied volatility, technical trading, data snooping and market efficiency
Author: Psaradellis, Ioannis
ISNI:       0000 0004 6496 5856
Awarding Body: University of Liverpool
Current Institution: University of Liverpool
Date of Award: 2017
Availability of Full Text:
Access from EThOS:
Access from Institution:
The first essay concentrates on the modelling and trading of three daily market implied volatility indices issued on the Chicago Board Options Exchange (CBOE) using evolving combinations of prominent autoregressive and emerging heuristics models, with the aim of introducing an algorithm that provides a better approximation of the most popular U.S. volatility indices than those that have already been presented in the literature and determining whether there is the ability to produce profitable trading strategies out-of-sample. A heterogeneous autoregressive process (HAR) is combined with a genetic algorithm–support vector regression (GASVR) model in two hybrid algorithms. The algorithms' statistical performances are benchmarked against the best forecasters on the VIX, VXN and VXD volatility indices. The trading performances of the forecasts are evaluated through a trading simulation based on VIX and VXN futures contracts, as well as on the VXZ exchange traded note based on the S&P 500 VIX mid-term futures index. Our findings indicate the existence of strong nonlinearities in all indices examined, while the GASVR algorithm improves the statistical significance of the HAR processes. The trading performances of the hybrid models reveal the possibility of economically significant profits. This second essay investigates the debatable success of technical trading rules, through the years, on the trending energy market of crude oil. In particular, the large universe of 7846 trading rules proposed by Sullivan et al., (1999), divided into five families (filter rules, moving averages, support and resistance rules, channel breakouts, and on-balance volume averages), is applied to the daily prices of West Texas Intermediate (WTI) light, sweet crude oil futures as well as the United States Oil (USO) fund, from 2006 onwards. We employ the k-familywise error rate (kFWER) and false discovery rate (FDR) techniques proposed by Romano and Wolf (2007) and Bajgrowicz and Scaillet (2012) respectively, accounting for data snooping in order to identify significantly profitable trading strategies. Our findings explain that there is no persistent nature in rules performance, contrary to the in-sample outstanding results, although tiny profits can be achieved in some periods. Overall, our results seem to be in favour of the adaptive market hypothesis. The third essay examines technical trading rules performance on the statistical arbitrage investment strategy, pairs trading, using daily data over the period 1990- 2016 for 15 commodity, equity and "famous" currency pairs. Adopting the false discovery rate test of Barras et al., (2010) to control for data snooping bias and exercising 18,412 technical trading rules, we find evidence of significant predictability and excess profitability, especially for commodity spreads, where the best performing strategy generates an annualized mean excess return of 17.6%. In addition, we perform an out-of-sample analysis to cross-validate our results in different subperiods. We find that whilst the profitability of rules based on technical analysis exhibits a downward trend over the sample, the opportunities for pairs trading remains has increased in certain cases.
Supervisor: Pantelous, A. ; Laws, J. Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral