Performance of a Nonparametric Multivariate Nearest Neighbor Model in the Prediction of Stock Index Returns

Abstract

This paper investigates the predictability of stock index returns using a nonparametric multivariate nearest neighbor model. A cross validation method that minimizes the mean square error is used to determine the embedding dimensions and the number of neighbors optimally. The performance of the proposed model is demonstrated using the KOSPI composite index and 16 industry indexes.