Financial News Stock Market
The complex behavior of financial markets emerges from decisions made by many traders. Here, we exploit a large corpus of daily print issues of the Financial Times from 2nd January 2007 until 31st December 2012 to quantify the relationship between decisions taken in financial markets and developments in financial news. We find a positive correlation between the daily number of mentions of a company in the Financial Times and the daily transaction volume of a company's stock both on the day before the news is released and on the same day as the news is released. Our results provide quantitative support for the suggestion that movements in financial markets and movements in financial news are intrinsically interlinked.
financial news stock market
The movements of stock markets impact the lives of many individuals, within the financial sector and far beyond. Obvious benefits therefore lie in an improved understanding of the behavior of this complex system. Research towards this goal has been fuelled by the vast amount of data on financial transactions recorded at exchanges, with increasing numbers of studies in complex systems science aiming to analyze1,2,3,4,5,6,7 and model stock market behavior8,9,10,11,12.
Recent research has sought to investigate whether data on what information users seek online can provide insight into market movements. Preis, Reith and Stanley provided initial evidence of a link between online searches and financial market behavior, describing a correlation between the weekly number of Google searches on a company name and weekly cumulative transaction volume of the corresponding company's stock23. Preis, Moat and Stanley built on this result, demonstrating that changes in Google query volume for search terms related to finance could be interpreted as early warning signs of stock market moves24. Moat et al. showed that data on views of Wikipedia pages can also be related to market movements, providing evidence of increases in views of financially related Wikipedia pages before stock market falls25. Evidence has also been provided that Google Trends data can be used to measure the risk of investment in a stock26.
Traders may however not only receive information through explicit attempts to search for information online, but by passively or actively receiving news broadcast by large financial news outlets. Equally, the actions of traders may lead to events which are described by the financial news. In this study, we seek to quantify the relationship between movements in financial news and movements in financial markets by exploiting a corpus of six years of financial news.
To examine the relationship between financial news and market behavior, we analyze a corpus of daily issues of the Financial Times from 2nd January 2007 to 31st December 2012. Details of how the corpus was retrieved and preprocessed are provided in the Supplementary Information.
A total of 891,171 different words occur throughout the Financial Times corpus. We begin our investigation of the relationship between financial news and financial market movements by focusing on occurrences of the names of the 31 companies that were listed in the Dow Jones Industrial Average (DJIA) between 2nd January 2008 and 31st December 2012, a period for which we have transaction volume and price data for the DJIA components. At any one time, the DJIA consists of 30 companies. However, Travelers replaced Citigroup in the DJIA during the period of our analysis. In the calculations reported, we consider stock data and news data for both of these companies. Full details of the company names used in the corpus analysis are provided in the Supplementary Information (Table S1).
We investigate the relationship between interest in a company in the news and interest in a company in the stock markets. Stocks for companies listed in the DJIA are traded at the New York Stock Exchange (NYSE), open between 9:30 am and 4 pm New York time (for most of the year, 2:30 pm to 9 pm London time). We carry out this analysis and all following analyses for trading days only, excluding all weekends and bank holidays.
We extend this analysis to all 31 Dow Jones companies from this period. For each company, we calculate the Spearman's rank correlation between the daily number of mentions of a company's name in the Financial Times and the transaction volume of the corresponding company's stocks (Figure 3). We analyze the distribution of Spearman's rank correlation coefficients for all 31 companies. Whilst the strongest correlation is found for Bank of America, we find that overall, the correlation coefficients are significantly higher than zero (median correlation coefficient = 0.074; mean correlation coefficient = 0.100, W = 450, p
We examine whether there is a similar link between the daily number of mentions of a company's name and the daily absolute return of the corresponding company's stocks. The absolute return indicates how much a stock price has changed, regardless of its direction. As a greater volume of trading is known to be correlated with greater movements in the price of a company's stock, it would be reasonable to expect the relationship between news and absolute return to be similar to the relationship we find between news and transaction volume.
The daily return is defined as the natural logarithm of the ratio of the closing price of a given day to the closing price from the previous day. We compute the absolute daily return for each of the 31 companies by taking the absolute values of the daily returns and calculate the Spearman's rank correlation between the daily number of mentions of a company and the company's daily absolute return (Figure 4). Again, we find that across all 31 companies, the correlation coefficients are significantly higher than zero (median correlation coefficient = 0.040; mean correlation coefficient = 0.047; W = 408, p = 0.0017, Wilcoxon signed rank test). Our results therefore suggest that greater interest in a company in the news is also related to greater movements in the company's stock price in the markets.
We investigate whether a relationship also exists between interest in a company in the news and movement in a company's stock price when direction of movement is taken into account. We calculate the Spearman's rank correlation between the daily number of mentions of a company and the daily return of a company's stocks (Figure 5) and find that here, the correlation coefficients are not significantly different to zero (median correlation coefficient = 0.000, mean correlation coefficient = 0.002, W = 262, p = 0.784, Wilcoxon signed rank test). In other words, our analysis so far provides no evidence that interest in a company in the news is correlated with company stock price movements when direction of movement is considered.
In summary, we find evidence for a relationship between interest in a company in the news on a given day and both the volume of trading and size of price change for a company's stocks on the same day. We find no evidence for a relationship between interest in a company in the news on a given day and price change for a company's stocks on the same day when direction of this change is taken into account.
However, whilst we are linking news released at 5 am London time on a given day with trading in a market much later in the day, between 9:30 am and 4 pm New York time, our current analyses do not allow us to draw strong conclusions about whether news influences the markets, or the markets influence the news. To gain some understanding of the directionality of this relationship, we extend this analysis by considering the relationship between mentions of a company in the news on a given day and transaction volume for a company on the three days beforehand and the three days afterwards (Figure 6).
We use six years of daily print issues of the Financial Times to quantify the relationship between decisions taken in financial markets and developments in financial news. We analyze mentions of the companies that form the Dow Jones Industrial Average and find that a greater number of mentions of a company in the news on a given morning corresponds to a greater volume of trading for that company during a given day, as well as a greater change in price for a company's stocks. Our analyses also uncover a link between the volume of trading for a company and the number of mentions of company in the news on the next day. Our current analysis provides no evidence of a relationship between the number of mentions of a company in the morning's news and the change in price for a company's stocks when direction of price movement is taken into account.
The results we present here are consistent with the hypothesis that movements in the news and movements in the markets may exert a mutual influence upon each other. Future analyses building on this work will seek to provide further insight into the direction and causality of the relationship between financial news and market movements.
We compile a list of common names used for the 31 companies listed in the Dow Jones Industrial Average (DJIA) between 2nd January 2008 and 31st December 2012. At any one time, the Dow Jones Industrial Average (DJIA) is derived from the stock prices of 30 companies. On 8th June 2009, during the period of our study, Travelers replaced Citigroup in the DJIA. In our analyses, we consider stock data and news data for both of these companies throughout the whole period.
Equity markets began the year with more of a risk-on tone, as traditionally cyclical areas like financials, materials, small-cap stocks, and even more speculative parts of the market outperformed. Over the past month, stock-market leadership has shifted to reflect more defensive positioning, with defensive sectors like health care and consumer staples outperforming, alongside parts of growth and technology. Growth sectors have been viewed recently as somewhat resilient in the midst of the banking turmoil, especially those companies with strong financial positions and healthy free cash flow. These sectors also have done well when rates move lower, which has been the case in recent weeks too. 041b061a72