VR Logo

Not by Numbers Alone

Data is great, but it is just another tool to provide insight with limitations

Data is supposed to be at the heart of the modern investment process. In fact, over the last two decades, Value Research has done more to promote data-centric and data-derived investment strategies than anyone else in India. Till a few years back, we were doing this mostly for mutual fund investing but with the launch of Wealth Insight eight years ago we have also brought our unique approach of data-centric strategies to equity investing. Recently, we have extended this to our website with a fabulous new stock snapshot. Search for any company from the search box on the top of every page to try it out. However, there's data, and there's DATA. Over the last few years, the ever-broader availability of lots of different type of data has resulted in a new kind of a problem. Investors and researchers worship at the altar of data of all sorts that could possibly have an impact on their investments.

But it's not just business and economic numbers. The domains from which supposedly relevant data could possibly originate are indeed numerous. It could be demographic trends, or weather phenomena like El Nino, and even the results of global sports events, as we are being told nowadays. Not just that, a recent article in The Economist claimed that cold weather helps the stock markets, after noting that hot weather was already known to do so!

No matter how interesting they may be, poring over temperature and rainfall graphs of data over decades will not help anyone make better investment or even business decisions. In fact, the abundance of data of all sorts actually creates a problem. It allows data to be found that can support almost any conclusion, depending on the situaution or the need. This is quite obvious when you read or watch some 'research' based on such data. The analyst has clearly started with a conclusion and then gone about hunting down data that can prove the conclusion in a manner convenient to the research.

A close relative of this is the activity of looking for cause-and-effect in things that just happen to be correlated. While this is far more common in health and science articles, but business and investing are not immune to it. In fact, there's a website, tylervigen.com, that is dedicated to spurious correlations. For instance, did you know that over the last 15 years, there has been a strong correlation between the number of people who drown in swimming pools every year and the number of films that the Hollywood actor Nicholas Cage has acted in during that year? Cause and effect? Perhaps not. But then, is the connection between cold weather and stock prices more plausible, or does it merely appear to be so?

So does that mean that investors should cut down on using data to study stocks? Not quite. However, the availability of excess data does mean that we have to take care to distinguish between what is useless (even if its interesting) and what is actually valuable. What is of lasting and real value is data and knowledge which helps us judge whether a company has a quality management on the basis of long track record.

In the long run, it matters less what kind of rain falls on the performance of a stock than what is the quality of the management response. You just need to use data to identify quality companies, based on factors not just governed by data and then assemble them into a large and diversified portfolio so that the negative effects of company and sector-specific events can be evened out.

Data is great, but it's just another tool. It's something that can be used to provide insight, but like all tools, it can do more harm than good if not used expertly.