Investing Report

...

Based on our model, we select the Shanghai and Shenzhen 300 index to represent the market portfolio and collect its daily rate of return from June 8, 2014 to June 8, 2017. Besides, according to the Wind industry classification, we collect the daily rate of return of the 10 sections---Material, Real estate, Industry, Infrastructure, Finance, Energy, Agriculture, Daily consumption, Information Technology (IT) and Health Care--- from June 8, 2014 to June 8, 2017. (Data source: Wind). By using the ARCH (1) model, and we get the following results.

2.2.1 Result (Arch Model)

Table 1. Arch Results in Each Section

Firm Number

[pic 25]

[pic 26]

[pic 27]

Material

253

0.012

0.299

1.010

(0.000191)***

(0.0249)***

(0.335)***

Real estate

135

0.013

0.223

-1.953

(0.000360)***

(0.0309)***

(0.447)***

Industry

119

0.012

0.252

0.183

(0.000238)***

(0.0211)***

(0.258)

Infrastructure

147

0.012

0.279

1.260

(0.000298)***

(0.0311)***

(0.408)***

Finance

312

0.010

0.431

-3.620

(0.000203)***

(0.0192)***

(0.243)***

Energy

68

0.012

0.425

-4.193

(0.000411)***

(0.0276)***

(0.400)***

Agriculture

41

0.013

0.344

0.455

(0.000448)***

(0.0352)***

(0.463)

Consumption

143

0.013

0.305

0.969

(1.08e-06)***

(0.0295)***

(0.409)**

IT

192

0.014

0.499

-1.197

(0.000480)***

(0.0364)***

(0.478)**

Health care

147

0.003

0.036

-11.19

(0.00111)**

0.0823

(1.061)***

Above is the summarized result from each section in China’s stock market. For Real estate, Finance, Energy, IT and Health care sections, the existence of herding effect is statistically significant and investors are likely to behave by following the lead without rational thinking. It might be because the information in such sections is more difficult to obtain and individual investors have to speculate by following the “big guys”. The herding effect in Health care section is the most typical, which means the investors are more concerned about what the big shareholders do and they tend to keep consistent with them. For Real estate and IT Industry, although the herding effect is significant, the coefficient is small. The reason may be that the big shareholders are not willing to reveal their information or they tend to tell the public in a later time instead of immediately. Thus a lagging effect may arise and needs to be considered.

In Material, Infrastructure and Daily consumption sections, the herding effect does not exist. As for Agriculture, it is hard to tell whether the herding effect exists or not.

2.3 OLS Model

After figuring out whether herding effect exists in different industry, we want to find out whether the stock returns will be affected if there is herding effect in its industry. Thus, on the basis of the CAPM model, we add a dummy variable that represents the herding effects and build the following OLS model:

[pic 28]

We use the nearly annual rate of return (2016~2017) of the stocks which we used to analyze the herding effect as the dependent variable Ri and used CAPM model to calculate the expected rate of return as the first independent variable. As