Independent Variable Selection: An Application of Independent Component Analysis to Forecasting a Stock Index
Forecasts of financial time series requires the use of a possibly large set of input (explanatory) variables drawn form a very large set of potential inputs. Selection of a meaningful and useful subset of input variables is a formidable task. How to find a reasonable transformation for a large set of multivariate data is a very common problem in many areas of science. We propose to use a technique called Independent Component Analysis (ICA) to extract the independent components (ICs) from monthly time series on a wide range of economic variables. This procedure will reduce the number of explanatory variables by reducing the set of financial and economic information to a much smaller subset of ICs which hopefully will capture most of the useful information. Removal of the random elements in each of the sets of economic data should make it much easier to identify relationships between the ICs and the stock indexes. Properly estimated ICs are independent of each other. We will then use the ICs from the explanatory variable data sets to perform and test forecasts of the S & P 500 stock index using neural network procedures. Numerous studies such as have shown neural networks to be very useful in nonlinear forecasting. IC analysis has been employed in relatively few applications to finance. Kiviluoto and Oja (1998) use IC analysis in an application to parallel cash flow time series. Black and Weigand (1997) use IC analysis to extract estimates of the structure from a set of common stock returns. We feel that this research will contribute to the identification and understanding of the major economic factors affecting stock prices.
|Year of publication:||
|Authors:||Cichocki, A ; Stansell, S R ; Leonowicz, Z ; Buck, J|
|Type of publication:||Other|
Saved in favorites
Similar items by subject
Find similar items by using search terms and synonyms from our Thesaurus for Economics (STW).