|
on Computational Economics |
|
Issue of 2006‒11‒18
four papers chosen by |
| By: | Francesco Bertoluzzo (Consorzio Venezia Ricerche); Marco Corazza (Department of Applied Mathematics, University of Venice) |
| Abstract: | In this paper we propose a financial trading system whose trading strategy is developed by means of an artificial neural network approach based on a learning algorithm of recurrent reinforcement type. In general terms, this kind of approach consists: first, in directly specifying a trading policy based on some predetermined investorâs measure of profitability; second, in directly setting the financial trading system while using it. In particular, with respect to the prominent literature, in this contribution: first, we take into account as measure of profitability the reciprocal of the returns weighted direction symmetry index instead of the wide-spread Sharpe ratio; second, we obtain the differential version of the measure of profitability we consider, and obtain all the related learning relationships; third, we propose a simple procedure for the management of drawdown-like phenomena; finally, we apply our financial trading approach to some of the most prominent assets of the Italian stock market. |
| Keywords: | Financial trading system, recurrent reinforcement learning, no-hidden-layer perceptron model, returns weighted directional symmetry measure, gradient ascent technique, Italian stock market. |
| JEL: | C45 C61 C63 G31 |
| Date: | 2006–10 |
| URL: | http://d.repec.org/n?u=RePEc:vnm:wpaper:141&r=cmp |
| By: | Diana Barro (Department of Applied Mathematics, University of Venice); Antonella Basso (Department of Applied Mathematics, University of Venice) |
| Abstract: | This contribution studies the effects of credit contagion on the credit risk of a portfolio of bank loans. To this aim we introduce a model that takes into account the counterparty risk in a network of interdependent firms that describes the presence of business relations among different firms. The location of the firms is simulated with probabilities computed using an entropy spatial interaction model. By means of a wide simulation analysis we use the model proposed to study the effects of default contagion on the loss distribution of a portfolio. |
| Keywords: | credit risk, bank loan portfolios, contagion models, entropy spatial models |
| JEL: | G33 G21 C15 |
| Date: | 2006 |
| URL: | http://d.repec.org/n?u=RePEc:vnm:wpaper:143&r=cmp |
| By: | Eleni Constantinou (Department of Accounting and Finance, The Philips College, 4-6 Lamias Street, CY-2100, Nicosia,); Robert Georgiades (Department of Accounting and Finance, The Philips College, 4-6 Lamias Street, CY-2100, Nicosia,); Avo Kazandjian (Department of Business Studies, The Philips College, 4-6 Lamias Street, CY-2100, Nicosia, Cyprus.); George Kouretas (Department of Economics, University of Crete, Greece) |
| Abstract: | This paper provides an analysis of regime switching in volatility and out-of-sample forecasting of the Cyprus Stock Exchange using daily data for the period 1996-2002. We first model volatility regime switching within a univariate Markov-Switching framework. Modelling stock returns within this context can be motivated by the fact that the change in regime should be considered as a random event and not predictable. The results show that linearity is rejected in favour of a MS specification, which forms statistically an adequate representation of the data. Two regimes are implied by the model; the high volatility regime and the low volatility one and they provide quite accurately the state of volatility associated with the presence of a rational bubble in the capital market of Cyprus. Another implication is that there is evidence of regime clustering. We then provide out-of-sample forecasts of the CSE daily returns using two competing non-linear models, the univariate Markov Switching model and the Artificial Neural Network Model. The comparison of the out-of-sample forecasts is done on the basis of forecast accuracy, using the Diebold and Mariano (1995) test and forecast encompassing, using the Clements and Hendry (1998) test. The results suggest that both non-linear models equivalent in forecasting accuracy and forecasting encompassing and therefore on forecasting performance. |
| Keywords: | Regime switching, artificial neural networks, stock returns, forecast |
| JEL: | G |
| Date: | 2005–01 |
| URL: | http://d.repec.org/n?u=RePEc:crt:wpaper:0502&r=cmp |
| By: | Paolo Pellizzari (Department of Applied Mathematics, University of Venice); Arianna Dal Forno (Department of Applied Mathematics, University of Venice) |
| Abstract: | We compare price dynamics of different market protocols (batch auction, continuous double auction and dealership) in an agent-based artificial exchange. In order to distinguish the effects of market architectures alone, we use a controlled environment where allocative and informational issues are neglected and agents do not optimize or learn. Hence, we rule out the possibility that the behavior of traders drives the price dynamics. Aiming to compare price stability and execution quality in broad sense, we analyze standard deviation, excess kurtosis, tail exponent of returns, volume, perceived gain by traders and bid-ask spread. Overall, a dealership market appears to be the best candidate, generating low volume and volatility, virtually no excess kurtosis and high perceived gain. |
| Keywords: | Agent-based models, artificial markets, comparison of market protocols. |
| JEL: | N22 D44 C15 |
| Date: | 2006 |
| URL: | http://d.repec.org/n?u=RePEc:vnm:wpaper:140&r=cmp |