Evaluating trade classification algorithms: Bulk volume classification versus the tick rule and the Lee-Ready algorithm

被引：28

作者：

Chakrabarty, Bidisha ^{[1
]}

Pascual, Roberto ^{[2
]}

Shkilko, Andriy ^{[3
]}

机构：

[1] St Louis Univ, St Louis, MO 63103 USA

[2] Univ Balearic Isl, Balearic Islands, Spain

[3] Wilfrid Laurier Univ, Waterloo, ON N2L 3C5, Canada

来源：

JOURNAL OF FINANCIAL MARKETS | 2015年 / 25卷

基金：

加拿大创新基金会;

关键词：

Trade classification; Bulk volume classification; Tick rule; Lee and Ready; VPIN; STOCK RETURNS; LIQUIDITY; INFORMATION; DIRECTION;

D O I：

10.1016/j.finmar.2015.06.001

中图分类号：

F8 [财政、金融];

学科分类号：

0202 ;

摘要：

We compare the accuracy of the bulk volume classification (BVC) to that of the tick rule (TR) and the Lee-Ready (LR) algorithm for a large sample of equities. TR and LR produce significantly better classifications than the BVC This result applies to stocks of all sizes, including the most frequently traded. Iteratively optimizing the BVC improves its performance, but the conventional rules still outperform. TR and LR produce more accurate estimates of the volume-synchronized probability of informed trading. Order imbalances computed using TR and LR are comparable to those computed using the BVC in explaining returns, liquidity, and trading costs. (C) 2015 Elsevier B.V. All rights reserved.

引用

页码：52 / 79

页数：28

共 26 条

[1] Illiquidity and stock returns: cross-section and time-series effects [J].