Рациональность инвесторов при оценке "критической массы" на фондовом рынке
The question about possibilities to use Twitter users’ moods to increase accuracy of stock price movement prediction draws attention of many researchers. In this paper we examine the possibility of analyzing Twitter users’ mood to improve accuracy of predictions for Gold and Silver stock market prices. We used a lexicon-based approach to categorize the mood of users expressed in Twitter posts and to analyze 755 million tweets downloaded from February 13, 2013 to September 29, 2013. As forecasting technique, we select Support Vector Machines (SVM), which have shown the best performance. Results of SVM application to prediction the stock market prices for Gold and Silver are discussed.
Both, business and academic communities agree that corporate news do affect the company market value. Empirical data shows that once released in the open, corporate news often lead to a rather predictable investor reaction. This investor reaction depends on a great number o factors: whether the news is good or bad, what type of corporate event has lead to the news, how broad is the analyst coverage of the company, what were the preceding company and analyst forecasts, prevailing stock market dynamics at the time, type of company shares, and a dozen of other factors. In our work, we attempted to put together disjoint empirical data, filter out the most significant common factors, determine their influence on the company value, and come up with a coherent big picture. Thereby we have developed a conceptual model that describes what kind of news and under what conditions will influence the company stock price this way or the other. We also propose a qualitative methodology for estimating the influence of news on the stock price. Our model and methodology are meant to help companies to better anticipate market reaction to their corporate announcements, and therefore correct possible negative impact leading to overall more efficient value based management.
In this paper we study the existence of a speculative bubble in the prices of the Russian telecommunications companies in the late 1990-s. In the study we use the regime-switching-type of econometric test that diagnoses the explosive pattern in the stock prices. Tests that compare series of prices and dividends are not applicable because most of the Russian quoted telecom companies did not pay dividends at that time. The tests did not reject the hypothesis of the existence of the bubble in the prices of shares of all Russian telecommunication companies except one and in the index of the Russian telecoms. The existence of the bubble was confirmed by the presence of its indirect indicators, namely rapid price growth and abnormal cumulative return before the price peak, too optimistic perception of news related to telecoms, deep correction and lengthy recovery after the crash. From 1998 to 2000 the prices of the Russian telecom companies grew in line with the Russian stock market, the correlation coefficient being 0,95. The telecoms index broke away from the general stock market only in the 1Q of 2000. We also tested whether the bubble in the Russian telecommunication companies stock market was a result of contagion from the NASDAQ market or it was caused by the revival of the Russian stock market after the 1998 crisis. The same was studied with respect to the crash of the bubble. We analyzed the dynamics of the correlation coefficients between the markets. This methodology implies that the markets become more interdependent and the contagion takes place if the correlation increases after the shock. No contagion effect was diagnosed. The indirect indication of the fact that no contagion took place is the date of the peak of the Russian telecoms index. It was reached 19 days after the NASDAQ peak. The speculative bubble on the Russian telecommunication stock market was determined by the events in the Russian stock market, but was influenced by news and attitudes toward telecommunications stocks in the USA.