A Nonparametric Kernel Regression Method for the Recognition of Visual Technical Patterns in China´s Stock Market

Author

Wang, Zhigang ; Zeng, Yong ; Li, Ping

Author_Institution

Sch. of Manage. & Econ., Univ. of Electron. Sci. & Technol. of China, Chengdu, China

fYear

2010

fDate

13-15 Aug. 2010

Firstpage

296

Lastpage

300

Abstract

This paper investigates the informativeness of some popular visual technical patterns among technical practitioners in China´s stock market, which are quantitatively defined and automatically recognized using a nonparametric kernel regression method. By comparing the unconditional empirical distribution of daily returns to the distribution of the returns that follow these technical patterns, we find that most of the patterns can provide incremental information that may be used to forecast further prices changes. However, only two of eight technical patterns can not generate significant excess trading profits after risk adjustment.

Keywords

economic forecasting; nonparametric statistics; pattern recognition; regression analysis; security of data; stock markets; China stock market; nonparametric Kernel regression method; prices change forecasting; risk adjustment; unconditional empirical distribution; visual technical pattern recognition; Companies; Finance; Kernel; Pattern recognition; Smoothing methods; Stock markets; Visualization; excess return; kernel regression; pattern recognition; technical patterns;

fLanguage

English

Publisher

ieee

Conference_Titel

Business Intelligence and Financial Engineering (BIFE), 2010 Third International Conference on

Conference_Location

Hong Kong

Print_ISBN

978-1-4244-7575-9

Type

conf

DOI

10.1109/BIFE.2010.76

Filename

5621837