 Evolutionary Strategies vs. Neural Networks; New Evidence from Taiwan on the Divisia Index DebateGraham Kendall, Jane Binner and Alicia Gazely No 176, Computing in Economics and Finance 2001 from Society for Computational Economics Abstract: In recent years the relationship between ÎmoneyÌ and the macroeconomy has assumed prominence in the academic literature and in Central Banks circles. Although some Central Bankers have stated that they have formally abandoned the notion of using monetary aggregates as indicators of the impact of their policies on the economy, research into the link between some kind of monetary aggregate and the price level is still prevalent. Attention is increasingly turning to the method of aggregation employed in the construction of monetary indices. The most sophisticated index number used thus far relies upon the formulation devised by Divisia. The construction has it roots firmly based in microeconomic aggregation theory and statistical index number theory. Our hypothesis is that measures of money constructed using the Divisia index number formulation are superior indicators of monetary conditions when compared to their simple sum counterparts. Our hypothesis is reinforced by a growing body of evidence from empirical studies around the world which demonstrate that weighted index number measures may be able to overcome the drawbacks of the simple sum, provided the underlying economic weak separability and linear homogeneity assumptions are satisfied. Ultimately, such evidence could reinstate monetary targeting as an acceptable method of macroeconomic control, including price regulation. We offer an exploratory study of the relevance of the Divisia monetary aggregate for Taiwan over the period 1978 to date. We adopt the principles of Ford (1997) by allowing both for a period of gradual learning by individuals as they adapt to the financial changes and secondly by incorporating a mechanism to accommodate the changing perceptions of individuals to the increased productivity of money. Individuals are thus assumed to adjust their holdings of financial assets until the diffusion of financial liberalisation is complete. The novelty of this paper lies in the use of evolutionary strategies (ES) to examine TaiwanÌs recent experience of inflation. This is an unusual tool in this context and represents the first known application of its kind. Results are compared to those already produced for Taiwan using the Artificial Intelligence technique of neural networks to compare the explanatory power of both Divisia and simple sum measures of broad money as indicators of inflation. Evidence suggests that ESÌs compete favourably in small samples with the neural network results we have previously reported, with the added benefit that the architecture is simpler and thus easier to implement. In addition, the variety of parameters is reduced and thus further limits the decisions that have to be made when implementing the algorithm. Our preferred inflation forecasting model is achieved using 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. The paper concludes with a discussion of the promise of artificial intelligence techniques as new tools in the macroeconomic policy arena. Keywords: Evolutionary Strategies; Neural Networks; Divisia Money; Inflation Forecasting (search for similar items in EconPapers) There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it. Related works: Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text Persistent link: https://EconPapers.repec.org/RePEc:sce:scecf1:176 Access Statistics for this paper More papers in Computing in Economics and Finance 2001 from Society for Computational Economics Contact information at EDIRC. |
Is your work missing from RePEc? Here is how to contribute.
Questions or problems? Check the EconPapers FAQ or send mail to .
EconPapers is hosted by the
School of Business at Örebro University.