Search

We propose a statistical procedure to determine if a spatial structure observed in the data is generated by a deterministic (even chaotic) spatial process rather than by a stochastic process. This procedure can be used as a specification test. It is robust against non-linearity and non-stationarity and can complete the toolbox for testing the diagnosis as well. The advantages of the presented methods are high power, simplicity, and ease and ample applicability for tests to be conducted, provided that weak conditions are required. Herein, we conduct several simulations to evaluate the performance of our procedure on well-known spatial processes and in situations where standard tests for spatial autocorrelation fail to detect spatial dependence. Guidelines for using the technique are also provided. [ABSTRACT FROM AUTHOR]

This paper aims to introduce the concept of symbolic correlation integral SC that is extensively used in many scientific fields. The new correlation integral SC avoids the noisy parameter 𝜀 of the classical correlation integral, defined by Grassberger and Procaccia (1983) and extensively used for constructing correlation-integral-based statistics, as in the BDS test. Once the free parameter 𝜀 disappears, it is possible to construct a nonparametric powerful test for independence that can also be used as a diagnostic tool for model selection. The symbolic correlation integral is also extended to deal with multivariate models, and a test for causality is proposed as an example of the theoretical power of the new concept. With extensive Monte Carlo simulations, the paper shows the good size and power performance of symbolic correlation-integral-based tests. [ABSTRACT FROM AUTHOR]

In this article, we have proposed a simple diagnostic statistical procedure for testing the order of dependence of a spatial process. The proposed test is of nonparametric nature, and it is able to deal with potential nonlinear spatial dependencies. An added value is that from a methodological point of view, the new test is based on symbolic dynamics and hence on information theory. We characterized the behavior of a symbolic entropy measure in the presence of spatial dependencies of order higher than 1. The good power performance of the new method in detecting higher order spatial lags is notable and gives rise to an expectation that it may form a suitable basis for constructive specification searches. [ABSTRACT FROM AUTHOR]

In this letter the profitability of a simple trading rule based upon genetic algorithms has been investigated. The referred technical trading rule has been contrasted in four different sample periods of the Spanish stock market index known as IBEX-35. Results suggest that in general the profitability of the simple trading rule is superior to the buy-and-hold strategy. This conclusion is clearer in ‘bull’, ‘bear’ and ‘volatile’ market episodes. These results can be compared with those that apply artificial neural networks as simple trading strategies to the general index of Madrid. [ABSTRACT FROM AUTHOR]

This paper studies predictability and profitability of using neural networks (NN) in the Spanish security market. This is carried out through a hybrid approximation which entails evolving a genetic algorithm in order to obtain an optimal NN's architecture. To that end, (NNs) forecasts are transformed into a simple trading strategy, whose profitability is evaluated against a simple buy-and-hold strategy. [ABSTRACT FROM AUTHOR]

Estimation techniques based on Genetic Algorithms (GA) have been studied in the presence of cointegrated variables. Several applications of GA to time-series have ignored the fact that the equation estimated by GA might be spurious. In this line, in this study it is shown that: (1) GA robustly detects this kind of relationship when the process contains a linear cointegrated relationship, (2) estimated models provide real fitness instead of spurious fitness. The well known cointegrated relation between income and consumption is estimated using GA. [ABSTRACT FROM AUTHOR]

A traditional role of central banks has been to protect the purchasing power of money by keeping a roof on inflation. Recently, several central banks have explicitly assumed an inflation control target. This is the case, among others, of the European Central Bank (ECB) whose monetary policy goal is to keep an annual inflation rate below 2% over a medium-term horizon (ECB, 2001). In practice, inflation is often measured in terms of the Consumer Price Index (CPI). However, the headline inflation rate is not totally under the control of the ECB. In this study core inflation is measured by means of the structural VAR approach, in particular it is based on long-run restrictions over crucial macroeconomic variables: inflation, money and output. Core inflation is estimated and compared with the standard measured inflation. [ABSTRACT FROM AUTHOR]

One of the most popular tests for detecting non-linear structure is the BDS test. This test is based on the correlation integral, and some studies have proved the sensitivity of this concept in relation to the choice of one particular parameter: delay time. However, the BDS test considers fixed delay time equal to one. In this paper it is investigated whether a flexible choice of delay time might improve the capacity of the BDS test. It is shown that the BDS-G test (generalized BDS test, including delay time choice) can capture non-linearities that are undetected by the classical BDS test. Finally, the robustness of the BDS-G test in the presence of noise is analysed. [ABSTRACT FROM AUTHOR]