Raphael N. Markellos

EVALUATING PUBLIC TRANSPORT EFFICIENCY WITH NEURAL NETWORK MODELS

This paper is concerned with measuring performance of public transport services based on the concept of productive efficiency. A new nonparametric approach is proposed based on multi-layer perceptron neural networks (MLPs). The advantages and limitations of this approach are discussed and compared with those of mathematical programming and econometric techniques. The MLP is used, along with data envelopment analysis (DEA) and corrected least squares (COLS), to set out comparative annual efficiency measures for the London Underground, for the period 1970 to 1994. It is argued that the MLP approach is superior to traditionally applied techniques since it is both nonparametric and stochastic and offers greater flexibility. Finally, it is demonstrated that the proposed MLP efficiency analysis has important practical implications for decision making.

Seasonality in the Athens stock exchange

This paper studies calendar effects in the emerging Athens Stock Exchange. Rather than examining only basket indices, we analyse calendar effects for each of the constituent stocks of the Athens Stock Exchange General Index for the period from October 1986 to April 1997. In accordance with similar studies substantial evidence of ‘day-of-the week’, ‘monthly’, ‘trading month’ and ‘holiday’ effects are found. The intensity of these effects for various stocks on the basis of capitalization, beta coefficients and company type are examined. The results indicate that the calendar regularities vary significantly across the constituent shares of the General Index and that aggregation introduces a considerable bias in unravelling these regularities. Also, it is found that factors such as the beta coefficient and company type influence significantly the intensity of calendar effects.

OPTIMAL PRICE SETTING IN FIXED-ODDS BETTING MARKETS UNDER INFORMATION UNCERTAINTY

This paper develops a model of optimal pricing under information uncertainty for fixed‐odds betting markets. The model suggests that bookmakers require a premium for quoting the odds several days before an event. This premium reflects the uncertainty of public information that can be exploited by expert bettors. The model predicts that when bookmakers set optimal prices, expected returns to bettors increase as a monotonic function of winning probabilities. In this manner, an information‐based explanation is given for the celebrated favourite‐longshot bias in fixed‐odds. Using an extensive data‐set of football odds from two major European bookmakers, we estimate the probability of informed betting.

Wine Price Risk Management: International Diversification and Derivative Instruments

Variations in fine wine prices can be prominent and have widespread economic and financial implications. Although fine wine investments are dominated by French wines, we demonstrate that significant international diversification benefits exist for investors in Italian, Australian and Portuguese fine wines. This is important since we also find that diversification across varieties of French wine is not likely to be that effective. We propose the development of futures and options contracts on standardized wine price indices in order to enhance market completeness and to address the risk management needs of all market participants. Several popular continuous time processes are used to approximate empirically the dynamics of four fine wine price indices. On the basis of our results, we recommend appropriate equilibrium models for pricing fine wine futures and option contracts.

Diversification benefits in the smaller European stock markets

This paper examines the diversification benefits available to U.S. and Japanese investors over the period 1974-94 in seven of the smaller European stock markets (SESMs): Austria, Belgium, Greece, Holland, Ireland, Italy, and Spain. With reference to a simplified International CAPM that accommodates both contemporaneous and delayed information flows, we employ correlation, principal components, and cointegration analysis in studying monthly observations from national basket indices. The empirical evidence is conclusive in showing that the SESMs have behaved differently, at least since the October 1987 crash, with stronger contemporaneous interdependencies and integration between them and with the U.S. market. Cointegration analysis found no significant common trend shared between the SESMs and the U.S. and Japanese markets. We conclude that despite the increasing international integration there still exist opportunities for diversification investment in the smaller and less studied European stock markets.

Nonlinear modelling of European football scores using support vector machines

This article explores the linear and nonlinear forecastability of European football match scores using IX2 and Asian Handicap odds data from the English Premier league. To this end, we compare the performance of a Poisson count regression to that of a nonparametric Support Vector Machine (SVM) model. Our descriptive analysis of the odds and match outcomes indicates that these variables are strongly interrelated in a nonlinear fashion. An interesting finding is that the size of the Asian Handicap appears to be a significant predictor of both home and away team scores. The modelling results show that while the SVM is only marginally superior on the basis of statistical criteria, it manages to produce out-of-sample forecasts with much higher economic significance.

Sovereign debt markets in light of the shadow economy

We investigate the controversial role of the informal sector in the economy of 64 countries between 2003 and 2007 by focusing for the first time on the impact it has on sovereign debt markets. In addition to a standard ordered probit regression, we employ two nonparametric neural network modeling techniques in order to capture possible complex interactions between our variables. Results confirm our main hypothesis that the informal sector has significant adverse effects on credit ratings and lending costs. MLP neural networks offer the best fit to the data, followed by the RBF neural networks and probit regression, respectively. The results do not change with respect to the stage of economic development of a country and contradict views about the possibility of significant economic benefits arising from the informal sector. Our study has important implications, especially in the context of the ongoing sovereign debt crisis, since it suggests that a reduction in the informal sector of financially challenged countries is likely to help in relaxing credit risk concerns and cutting down lending costs. Finally, a decision tree analysis is used to exploit the inherent discreteness in the data and derive intuitive rules with respect to the level of the informal sector.

Corporate Real Estate Analysis:Evaluating Telecom Branch Efficiency in Greece

This article proposes productivity analysis for evaluating the relative efficiency in corporate real estate usage across decision-making units. Using data from the Greek Telecommunications Organization (GTO), we measure the productivity of 127 branches using the number of employees and the total area covered per building as inputs and the number of telephony access lines as outputs. We apply three nonparametric Data Envelopment Analysis (DEA) models assuming: Constant Returns to Scale (CRS), Variable Returns to Scale (VRS) and Slacks-Based Measures (SBM), respectively. We discuss how the proposed approach can provide real estate managers and analysts a multi-informational tool that allows the quantification of targets and may serve as a guide tool for the efficient employment of real estate assets.

Asset pricing dynamics

This paper is concerned with the issue of dynamics in financial data and asset pricing models such as the CAPM. A literature review in this area is undertaken and highlights the need for a modern time series econometric approach in asset pricing. Such an approach is discussed and deals with problems related to structural breaks and microstructures, dynamics in the mean and variance process, and non-stationary regressions and cointegration. An empirical application using UK stock market data demonstrates the merit of the proposed methodology in correcting market model regressions.

Diversification benefits in trading

This study argues that there may exist benefits in active portfolio management and trading other than the possibility of obtaining excess returns. The objective is not to attack the hypothesis that trading cannot produce (risk-adjusted) returns that are superior to passive investment strategies. What is suggested is that the combination of active and passive strategies can help considerably in diversifying investment positions. An empirical application using large samples of daily data on the Dow Jones Industrial Average (DJIA) and the Financial Times Institute of Actuaries 30 (FT30) indexes shows that simple market timing techniques, such as those used by Brock et al. (Journal of Finance, 47, 1731–64, 1992), Mills (International Journal of Finance and Economics, 2, 319–31, 1997) and Markellos (Applied Economics Letters, 6, 177–79, 1999), can be combined with buy-and-hold strategies to match the market return at a fraction of market risk. In accordance with the studies by Mills and Markellos, it is found that the behaviour of the data appears to have changed in recent years.