Seeking Alpha

How News Articles Can Really Help Us Invest In The Stock Market

by: EconReporter

Thanks to big data and machine learning, economists can use news articles to estimate investor or economic sentiment.

One popluar measure is the Economic Policy Uncertainty Index (EPU).

A new IMF research even tries to directly assess the impact of sentiment in the news on equity markets.

Information is power. It might also mean wealth in the financial market. Most investors know it is important to stay informed of the daily news flow, but many of us can't see the forest for the trees.

As John Maynard Keynes suggested, the stock market is like the beauty contest; It is the opinions of the voters, and the opinion of others' opinions, that counts. Which is why investor sentiment can be an important indicator of the future trend of the stock market. In recent years, because of the wide adoption of big data mining and machine learning technique in academic economics, more and more economists turn to newspapers to look for clues about the investor or economic sentiment.

A popular measure that is widely used by economists is the Economic Policy Uncertainty Index (NYSEARCA:EPU) developed by Scott Baker, Nicholas Bloom, and Stephen Davis. This is based on a scan of articles in leading newspapers looking for combinations of the following terms: “uncertainty” or “uncertain”; “economic” or “economy”; and policy terms like “Congress”, “deficit”, “Federal Reserve”, “legislation”, “regulation” or “white house”. The higher the ratio that news articles contain these keywords appear, the higher the EPU.

As shown in the graph below, EPU of the US spike in many major events for the financial markets.


The three economists find that an increase in EPU foreshadows a decline in economic growth and employment. They also find that the EPU predicts stock price volatility. As the number of large movements in the S&P 500 index (a daily change of 2.5% or more) has increased dramatically in recent years relative to the average since 1980, they attributed an increasingly large share of these large stock movements to policy-related events that can be captured by EPU.

Nonetheless, as shown below, the EPU of the US is not particularly high in 2018. This might be interpreted as an indication that policy-related factors are not a concern for the economy or the stock market.

Recently, in a new working paper from the IMF, four economists try to go one step further and directly assess the sentiment in the news and its impact on equity markets.

In the research "Media Sentiment and International Asset Prices," economists Samuel P. Fraiberger, Do Lee, Damien Puy and Romain Ranciere use the fraction of positive and negative words in each article in their 4.5 million Reuters article archive to capture the tone of news published, then construct a daily news-based sentiment index for 25 advanced and emerging countries between 1991 and 2015.

In their sample, about 60% of the corpus consists of local news, which only mentions one country and conveys rather specialized content. By contrast, the remaining 40% of the corpus contains articles discussing multiple countries. Typical multi-country news is an article, for example, about “Fears of Brazilian devaluation hit emerging markets,” which mentions multiple countries and their interrelations. The researchers use a lag of up to 8 days for the news sentiment index and try to predict the cumulative equity returns response for the next 20 days.

The figure below shows the effect of good news – measured by a one standard positive deviation in sentiment – on equity returns. The blue line is the local news sentiment, and the green one reflects global news sentiment.

Source: "Media Sentiment and International Asset Prices"

The cumulative response in the global stock market for an increase in "local good news" being roughly 5 basis points, the effect is temporary. This can be interpreted as a pure “sentiment” shocks affecting both advanced and emerging countries, during which the tone of local news affect investor sentiment and prices momentarily before returning to their fundamental values.

In sharp contrast, the global news sentiment shocks have a stronger and permanent effect on returns (green lines in the graph above). The total permanent impact is around 25 basis points roughly five times the size of the local sentiment shock. This long and lasting impact on world stock markets suggests that global news contain genuinely new information about fundamentals that is only slowly incorporated into stock prices;

Another analysis they perform is to check the relationship between the news sentiment index and equity flow. They find a very similar response like the one with equity prices. Although local news optimism attracts equity fund flows, it does so only temporarily. We estimate a statistically significant cumulative increase at the peak of 0.01% (blue lines in the graph below).

Optimism in global news (red lines in the graph below), on the other hand, generate a larger and permanent inflow of equity, which peaks after two weeks around 0.1%.

Source: "Media Sentiment and International Asset Prices"

In my opinion, this line of research is very promising. While it is still not possible to use such a technique to capture alpha, but these news articles mining based indicators can be further developed into some effective crisis warning signal.

Investors should look closer into the daily news flow, as this can be one way to avoid huge financial loss in a crisis.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.