ورود اطلاعات اینترنتی و نوسانات شاخص قیمت شرکت های کوچک و متوسط

This article employs the number of news appeared in Baidu News as a novel proxy for information arrival and investigates the validation of the Mixture of Distribution Hypothesis (MDH) using a sample of SME PRICE INDEX in China. The empirical results reveal a positive impact of internet information on the conditional volatility of stock returns. Compared with the prevailing proxies (trading volume and its adjustments), the volatility persistence is most decreased when this novel proxy is incorporated into the conditional variance equation of the GARCH model. Some tentative explanations are also given to expound the non-disappeared GARCH effects.

The volatility of the underlying asset prices has long been the focus of financial economics [1], [2] and [3], with the ultimate objective to clarify the potential causes and pricing mechanism. In accordance with Fama [4], the efficiency of the capital market is attributable in large part to how rapidly information is fully reflected into asset prices. Recent studies have concentrated on the relationship between the rate of information arrival and volatility by employing the ARCH and GARCH models. The Mixture of Distribution Hypothesis (MDH), which claims that the volatility of returns at a given interval is associated with the rate of information arrival, seems to be an appealing answer to explain why volatility is persistent [5], [6], [7] and [8]. Several previous studies have confirmed the validation of MDH. Lamoureux and Lastrapes [9] show a significant decrease in volatility persistence when raw trading volume is included in the conditional variance equation as the proxy for information arrival. After this, to avoid the nonstationary properties of the raw trading volume series, some forms of adjusted trading volume [10] and [3] as well as other proxies, including the number of firm-specific announcements [11] and transactions [12] are also applied to explain volatility persistence. Recent studies have shown evidence that the information provided by the internet is associated with economic activities [13], [14], [15], [16] and [17]. Among these open source information, search engine query data demonstrate a strong link between searching behavior and stock market performance [18], [19] and [20]. In this study, unlike previous proxies, we employ the number of news appeared in Baidu News1 as the direct proxy for information arrival and incorporate this novel variable into the conditional variance equation of the GARCH (1, 1) specification. The rationale for using this proxy is twofold. Firstly, trading volume (including its adjusted form) and transaction may not be the proper proxy for the information arrival. Because they cannot be assumed to be exogenous and trading activity is not solely driven by information [21] and [22]. Secondly, owing to the internet becoming the primary platform for information gathering,2 firm-specific announcement is not a sufficient representative proxy for the information arrival. The empirical results show a substantial reduction of the volatility persistence after the number Baidu News is incorporated into the conditional variance equation. The rest of the article is organized as follows. Section 2 describes the methodology. Section 3 presents the data and empirical results. Section 4 is the concluding remarks.

This article is deeply motivated by using internet data for economic research [15] and [26]. A direct proxy for information arrival is proposed to test the validation of MDH. Using a sample of SME PRICE INDEX in China, the empirical results reveal a positive and significant impact of the number of Baidu News on the conditional volatility of stock returns. Though, gaining some interesting findings, more work can be performed to investigate the influence of internet information on asset pricing. We leave this for future research.