Jane M. Binner

A comparison of linear forecasting models and neural networks: an application to Euro inflation and Euro Divisia

Linear models reach their limitations in applications with nonlinearities in the data. In this paper new empirical evidence is provided on the relative Euro inflation forecasting performance of linear and non-linear models. The well established and widely used univariate ARIMA and multivariate VAR models are used as linear forecasting models whereas neural networks (NN) are used as non-linear forecasting models. It is endeavoured to keep the level of subjectivity in the NN building process to a minimum in an attempt to exploit the full potentials of the NN. It is also investigated whether the historically poor performance of the theoretically superior measure of the monetary services flow, Divisia, relative to the traditional Simple Sum measure could be attributed to a certain extent to the evaluation of these indices within a linear framework. Results obtained suggest that non-linear models provide better within-sample and out-of-sample forecasts and linear models are simply a subset of them. The Divisia index also outperforms the Simple Sum index when evaluated in a non-linear framework.

VoiceYourView: collecting real-time feedback on the design of public spaces

This paper reports on VoiceYourView, a kind of intelligent kiosk, which uses speech recognition and natural language processing to gather the publics creative input on the public space designs. Over a six week period, VoiceYourView was deployed in a public space and 2000 design critiques were collected from 600 people. The paper shows that people are capable of providing creative input on their environment using unstructured speech or text and that a good proportion of these comments are actionable. The paper also investigates the use of public displays to auto-summarize comments left by the public so far. Although there is anecdotal evidence that this encourages participation, an experiment found that filtering comments (e.g., to display only positive responses) had no effect on what people had to say.

Forecasting exchange rates with linear and nonlinear models

In this paper, the exchange rate forecasting performance of neural network models are evaluated against the random walk, autoregressive moving average and generalised autoregressive conditional heteroskedasticity models. There are no guidelines available that can be used to choose the parameters of neural network models and therefore, the parameters are chosen according to what the researcher considers to be the best. Such an approach, however, implies that the risk of making bad decisions is extremely high, which could explain why in many studies, neural network models do not consistently perform better than their time series counterparts. In this paper, through extensive experimentation, the level of subjectivity in building neural network models is considerably reduced and therefore giving them a better chance of performing well. The results show that in general, neural network models perform better than the traditionally used time series models in forecasting exchange rates.

A composite leading indicator of the inflation cycle for the Euro area

We evaluate the performance of composite leading indicators of turning points of inflation in the Euro area, constructed by combining the techniques of Fourier analysis and Kalman filters with the National Bureau of Economic Research methodology. In addition, the study compares the empirical performance of Euro Simple Sum and Divisia monetary aggregates and provides a tentative answer to the issue of whether or not the UK should join the Euro area. Our findings suggest that, first, the cyclical pattern of the different composite leading indicators very closely reflect that of the inflation cycle for the Euro area; second, the empirical performance of the Euro Divisia is better than its Simple Sum counterpart and third, the UK is better out of the Euro area.

Financial innovation and Divisia monetary indices in Taiwan: a neural network approach

In this paper a weighted index measure of money using the ‘Divisia’ formulation is constructed for the Taiwan economy and its inflation forecasting potential is compared with that of its traditional simple sum counterpart. This research extends an earlier study by Gazely and Binner by examining the theory that rapid financial innovation, particularly during the financial liberalization of the 1980s, has been responsible for the poor performance of conventional simple sum monetary aggregates. The Divisia index is adjusted in two ways to allow for the major financial innovations that Taiwan has experienced since the 1970s. The technique of neural networks is used to allow a completely flexible mapping of the variables and a greater variety of functional form than is currently achievable using conventional econometric techniques. Results suggest that superior tracking of inflation is possible for networks that employ a Divisia M2 measure of money that has been adjusted to incorporate a learning mechanism to allow individuals to gradually alter their perceptions of the increased productivity of money. Divisia measures of money appear to offer advantages over their simple sum counter parts as macroeconomic indicators.

CO-EVOLVING NEURAL NETWORKS WITH EVOLUTIONARY STRATEGIES: A NEW APPLICATION TO DIVISIA MONEY

This work applies state-of-the-art artificial intelligence forecasting methods to provide new evidence of the comparative performance of statistically weighted Divisia indices vis-a-vis their simple sum counterparts in a simple inflation forecasting experiment. We develop a new approach that uses co-evolution (using neural networks and evolutionary strategies) as a predictive tool. This approach is simple to implement yet produces results that outperform stand-alone neural network predictions. Results suggest that superior tracking of inflation is possible for models that employ a Divisia M2 measure of money that has been adjusted to incorporate a learning mechanism to allow individuals to gradually alter their perceptions of the increased productivity of money. Divisia measures of money outperform their simple sum counterparts as macroeconomic indicators.

Tools for non-linear time series forecasting in economics - an empirical comparison of regime switching vector autoregressive models and recurrent neural networks

The purpose of this study is to contrast the forecasting performance of two non-linear models, a regime-switching vector autoregressive model (RS-VAR) and a recurrent neural network (RNN), to that of a linear benchmark VAR model. Our specific forecasting experiment is U.K. inflation and we utilize monthly data from 1969 to 2003. The RS-VAR and the RNN perform approximately on par over both monthly and annual forecast horizons. Both non-linear models perform significantly better than the VAR model.

Monetary models of exchange rates and sweep programs

Numerous studies find that monetary models of exchange rates cannot beat a random walk model. Such a finding, however, is not surprising given that such models are built upon money demand functions and traditional money demand functions appear to have broken down in many developed countries. In this article, we investigate whether using a more stable underlying money demand function results in improvements in forecasts of monetary models of exchange rates. More specifically, we use a sweep-adjusted measure of US monetary aggregate M1 which has been shown to have a more stable money demand function than the official M1 measure. The results suggest that the monetary models of exchange rates contain information about future movements of exchange rates, but the success of such models depends on the stability of money demand functions and the specifications of the models.

A Note On The Optimal Level Of Monetary Aggregation In The United Kingdom

Weak separability is a key admissibility property in the Divisia approach to monetary aggregation. We test groups of U.K. household sector monetary assets for weak separability using new data underlying the Bank of Englands benchmark revision of its household sector Divisia index. Nonparametric tests are used to identify four monetary asset groupings, which are weakly separable over all or almost all of the post-ERM period (1992:4-2005:1). We construct Divisia monetary aggregates for these four groupings and investigate their information content in two applications. The main findings are that Divisia money has direct effects on aggregate demand and that the growth rates of the nominal Divisia monetary aggregates Granger cause nominal output growth, but not inflation. (Less)

Mean-Variance vs. Full-Scale Optimization: Broad Evidence for the UK

In the Full-Scale Optimization approach the complete empirical financial return probability distribution is considered; and the utility maximizing solution is found through numerical optimization. Using a portfolio choice setting of three UK equity indices we identify several utility functions featuring loss aversion and prospect theory; under which Full-Scale Optimization is a substantially better approach than the mean-variance approach. As the equity indices have return distributions with small deviations from normality; the findings indicate much broader usefulness of Full-Scale Optimization than has earlier been shown. The results hold in and out of sample; and the performance improvements are given in terms of utility as well as certainty equivalents.