The Analysis of 52‐Week High Investing Strategy Based on Herding Behavior

• Published date: 01 March 2017
• Author: Chiao Yi Chang,Wen‐Hsiu Kuo,Hsiang‐Lan Chen
• Date: 01 March 2017
• DOI: http://doi.org/10.1111/irfi.12090

The Analysis of 52-Week High

Investing Strategy Based on

Herding Behavior

CHIAO YICHANG

†

,HSIANG-LAN CHEN

‡

AND WEN-HSIU KUO

†

†

Department of Insurance and Finance, National Taichung University of Science and

Technology, Taichung City, Taiwan and

‡

Department of Finance, National Kaohsiung First University of Science and

Technology, Kaohsiung, Taiwan

ABSTRACT

The present study develops zero-costing strategies that are based on the

52-week high and herding behavior. Proximity of the current price to the

52-week high and the level of herding behavior of individual/institutional

investors are the two criteria used to screen stocks. Because herding behavior

affects stocks that are associated with value-related beliefs that investors are

reluctant to revise, the level-of-herding criterion uses the 52-week high

strategy to improve profits. The present study examines strategy profits in

Taiwan, a market in which more than 70% of investors are individuals and

where the level of herding among individual investors is higher than that

for institutional investors. Empirical results found that profits earned using

zero-costing strategies, identified both using the 52-week high and herding,

were larger than those earned using only the 52-week high strategy.

Furthermore, stocks with values that were far from their 52-week high made

significant and positive profits through buy-herding and by shorting

sell-herding stocks.

1. INTRODUCTION

George and Hwang (2004) initially highlighted the critical importance of price

trends and the price maximum during the immediately preceding 52-week

period. Subsequently, numerous researchers have conducted studies into the

empirical evidence for this phenomenon in various markets, with similar results

obtained in Australia (Marshall and Cahan 2005), the U.K. (Burghof and

Prothmann 2011), 18 international stock markets (Lui et al. 2010), Iran

(Mansouri1 et al. 2013), and a basket of countries (Du 2008). However, while these

studies agree that a strategy of investing in stocks that are near their 52-week high

and that have short positions that are far from their 52-week high generated supe-

rior returns, the evidence provided by these experiments remains controversial.

* The authors appreciate the funding support from the Ministry of Science and Technology, R.O.C

under grant of NSC102-2410-H-025-008. We also appreciate the reviewers' constructive comments.

© 2016 International Review of Finance Ltd. 2016

International Review of Finance, 17:1, 2017: pp. 77–106

DOI: 10.1111/irfi.12090

Bornholt (2011) argued that the 52-week high effect is not as reliable as the mo-

mentum effect for both developed and emerging-market indices. Malin and

Bornholt (2010) even found negative returns in the market indices of emerging

markets. Alsubaie and Najand (2009) investigated the non-significant, positive

profits of the 52-week high effect in the Saudi stock market. They explained that

the differences in results between the Saudi stock market and developed financial

markets such as that in the U.S. might be attributed to differences in the diffusion

of information and to the overreactions of investors.

The currently large body of conflicting research related to the validity of the

52-week high effect opens new opportunities to focus attention on effects other

than the 52-week high. Therefore, the present study shifts attention from a focus

on the 52-week high effect to other variables that are related to volume and

explores the potential of these other variables to develop more effective

profit-driven strategies. The 52-week high relates to the concept of stock price,

which investors frequently use as an anchor. The 52-week high triggers investor

interest to invest, with the trend in investment volume then facilitating

discernment of the 52-week high effect. As the investment volume and price

level help define investor behavior, the present study adopts herding behavior,

arefined variable, to reflect buy orders and sell orders in order to define investor

behavior related to the 52-week high. Herding behavior relates strongly to

overreaction. Brown et al. (2014) argued that mutual-fund investors appear to

overreact when engaging in herd trading. According to Alsubaie and Najand

(2009), this overreaction provides a possible explanation for the 52-week high

effect that is frequently observed in emerging markets.

There are three reasons to designate herding behavior as a good proxy for

observing the actions of investors in relation to the 52-week high. First, the daily

herding variable described by Lakonishok et al. (1992) is calculated using

intra-day orders in order to reveal the power of buying/selling a specified stock

relative to that of the general buying/selling activity during the same time period.

Thus, this variable provides information that relates to the trading direction as

well as to the trading volume. Second, the concept of relativeness to all stocks

focuses greater attention on the status of the market than is gained from other

volume variables such as order imbalance. The implications of market status

include the level of active trades, which reflects current bull/bear market status

and economic conditions, among other variables. Conversely, order imbalance

indicates only the power represented by buying/selling an individual stock.

Third, the herding variable may be calculated according to investor type (e.g., in-

dividual investors, institutional investors) in order to more accurately adjust

estimates based on the current relative importance of each type.

Because herding is a short-term behavior in the daily base (Kremer and Nautz

2013), using daily data provides results that are more accurate and stable. The

literature recognizes that the length of the forming and holding period

significantly impacts profit. Another related topic, price momentum, has been

discussed in terms of daily frequency (Chang 2012). Most previous studies that

have investigated developed countries use monthly data. We executed a one-way

International Review of Finance

© 2016 International Review of Finance Ltd. 201678