Technical analysis, also known as “charting,” has received considerable attention for many decades. Much of the work of technical analysts has relied on the ability to recognize patterns when displayed pictorially. Some researchers have investigated the use of nonparametric methods to identify price patterns. Subjective bandwidth selection and boundary problems are two main issues when applying the Nadaraya–Watson kernel estimator to pattern recognition. To fill the gap, we propose a complete data-driven technical analysis algorithm with the application of a nonparametric local linear estimator. We incorporate trading volume together with trading price to define the patterns. Empirical implementation on S&P 500 Index stocks indicates that the proposed algorithm is very informative and promising. Available from: Price-Pattern Recognition Using a Local PolynomialRegression
Linear transformation models are a broad class of semiparametric regression models taking the proportional hazards model, proportional odds model and probit model as special cases. Although linear transformation models are widely used for analyzing right-censored survival data in the literature, their applications to current status data are limited.This paper proposes a unfied Bayesian estimation approach for regression analysis of current status data in all linear transformation models. Our proposed approach adopts monotone splines for modeling the unknown increasing functions in linear transformation models and uses shrinkage priors for the spline coefficients to allow spline basis selection and to avoid over-fitting. A novel slice sampler is proposed to facilitate the posterior computation and to allow one to estimate baseline function and regression parameters simultaneously. The proposed approach is generic for all linear transformation models and allows model selection. The method is illustrated through application to an epidemiological study of uterine fibroids.
Keywords: Unified approach, current status data, linear transformation model, monotone splines, semiparametric regression.