我们在2017年为美国华盛顿州大学写的一篇名为Market Pressure and Herding Phenomenon的essay是我们的essay范文，全文2500字，主要研究方向是A research on Chinese Stock Market，这类文章首先是要熟知相关大环境，之后再是分类去研究，契合题目的需求去深挖相关的资料，从而成文，首先分享introduction给大家。
Herding phenomenon has been a very intriguing topic in the stock market since 30 years ago, many scholars and economists tried dissecting the sources of the problem through theoretical researches. The term is mentioned when describing collective behaviors occur among investors under uneven or inadequate information. Because herding behaviors can be found among many entities and institutions, and are often associated with the stability and efficiency of the market, herding phenomenon can be the center of attention in the time of financial crisis (Chose et al, 1999).
The way to use open data like stock prices to detect herding phenomenon can be very critical. William (1995) made an extensive research on how to use dispersion of stock prices and its correlation to herding behavers. He believes that if there does exist herding behaviors in the market, most investors’ decision would comply to gossips in the market. Then there’s a reasonable conclusion that when there’s apparent herding behaviors in the market, the yield rate of individual stock won’t be too different from the yield rate of the market. Bikhchandani et al (1992) mentioned in their research that when there’s a larger volatility in market prices, there’s larger information uncertainty, and therefore larger potentials for herding behaviors.
Under the assumption above, we established a link between the dispersion of rate of return and the possibilities for herding phenomenon. We’ll use the dispersion of the rates of yields as a parameter assessing the degree of herding behaviors. By that, we use the standard deviation of individual stock returns to the portfolio average rate of returns, and this standard deviation, as a parameter for dispersion, can quantify the closeness of individual stock returns to collective portfolio returns. When the decision is entirely made by herding groups, all prices moves in the same volatility, and the standard deviation would be 0; And if there is one stock moves away from the portfolio, the degree of dispersion arises, and the standard deviation would be larger. We can detect the existence of herding behavior by comparing this parameter under different situation, both when there’s a large volatility in market prices and when there’s a small volatility where herding phenomenon’s hard to manifest.
Under the stress of market pressure, investors are more likely to abandon their own investment ideas and follow the trends, and therefore more likely to have herding behaviors. William (1995) suggested the following model:
D_t= α+ β_(1 ) C_t^L+ β_2 C_t^H+ ϵ_t
where α is a constant measuring the sample’s average dispersion degree which excludes the areas covered by the dummy variables; C_t^Land C_t^Hare dummy variables in market’s big rises and falls. And betas are the regression coefficients whose values will be the direct indicator for the existence of herding behaviors. And because there’s no such standard for market large volatility, when calculating the regression for the dispersion of daily rate of returns, we use 1% and 5% to define the market volatility. And this standard will confine the C_t^Land C_t^Hto the highest and lowest of 1% and 5% of the distribution function of rate of return.
We define C_t^Land C_t^Has follows:
Use as the market rate of return at time m, and〖 r〗^m (pl) and 〖 r〗^m (ph) as the p percentile for the distribution of market rate of return. When using 1% standard, pl = 0.01, and ph = 0.99; and when using 5% standard, pl – 0.05, and ph = 0.95.
If 〖 r〗^m (pl) ≤ 〖 r〗^m (ph), C_t^L= 1, else C_t^L = 0.
If 〖 r〗^m (pl) > 〖 r〗^m (ph), C_t^L = 1, else C_t^H = 0.
And if β_1< 0 and β_2< 0, then concludes there’s herding behaviors. If β_2>β_1, there are less herding behaviors when market’s higher rate of returns than lower rate of returns. And if β_2<β_1, vice versa.
In this essay, we discussed and analyzed the methodologies and its potential deficiencies in detecting herding phenomenon, and then we investigated the herding phenomenon in Chinese stock market and its causes, finally we provided few suggestions for the cause. Hopefully, it provides a comprehensive directory for herding behaviors in Chinese stock market and other less mature stock market alike.