Avoid the HiPPO! These three words can change the way anyone does business for the better.
Let's dive in.
But first, it’s time to confront a tough reality; data analytics is becoming an increasingly important part of the business world, including financial services. For financial professionals of all stripes, that means it’s now vitally important to understand how data analytics works, what it involves, and where it can be used.
Where Can Data Analytics Be Used?
Data analytics including predictive analytics (making data-driven predictions related to business issues) is useful for financial professionals in a broad variety of areas. Financial advisors and funds seeking institutional investors can use data analytics to optimize pricing and discounts for investors and clients. Investment banks can use data analytics to help corporate clients determine how the markets will react to particular corporate announcements (e.g. a merger) or to predict demand for securities offerings, hedge funds can use data analytics to value and assess complex illiquid securities, and compliance groups can help corporate execs understand risks from regulators using tools like the SEC’s new API measure or the any of the litany of other data related flagging tools.
Steps in Data Analytics
Understanding data analytics means starting by understanding the steps that all data analytics and business intelligences processes take.
There are five steps to any data analytics project; (1.) figure out what the question to be answered is, (2.) gather relevant data, (3.) clean and structure that data, (4.) run analysis and test hypotheses, (5.) make a decision.
Effective business analytics requires all five of these steps and it truly is a process. Any firm that sets out to use data on a one-off quick project is going to find themselves a victim of the garbage-in, garbage-out paradigm. Much of the consulting work that I do for firms is around using data to analyze business issues, and it’s very common for me to get a phone call from a potential client that has already tried to do a data analytics project quickly on their own often as a side task for one employee. It almost never works.
Nonetheless, data analytics is still crucial for businesses of all sizes today including financial advisors and portfolio managers.
Questions and Data Analytics
Anyone reading this can probably come up with questions they have about their own business that can be answered with data analytics. The biggest constraints on data analytics are not about having a question that requires numbers or math to answer – textual analysis can answer many non-mathematical questions. Rather the questions that data analytics can answer are those not based on a subjective opinion.
What is the probability of the merger between firm ABC and firm XYZ closing? What’s the probability that a firm beats its earnings estimates? What customers should I spend my time focusing on to maximize revenues and minimize customer defections? These are data analytics questions. What should I have for dinner tonight? That is not something data analytics can help with.
Where Does The Data Come From?
Once a question is determined, it’s time to gather data.
The biggest problem that firms run into when trying to do data analytics work, is that they are not sure where to get data from.
Fortunately there are numerous good tools out there to help with this issue today, and many are free or low cost.
There are many sources of free data such as the Federal Reserve’s FRED service and the Census Bureau’s Data Ferret. There are of course many paid data sources available as well from large firms like Thomson Reuters and Standard and Poors among many others.
What Is Involved in Analysis?
The process of cleaning up a data set and analyzing it requires some training and experience. Lay people can identify questions they are interested in answering, but answering those questions requires careful use of data. The most important tool in data analysis is the multiple regression often simply called regression analysis.
Regression analysis allows the analysis of a situation after controlling for all other factors involved. For example, what is the probability of a particular stock rising in value given the industry, earnings track record, investor base, market conditions, etc. etc.
Regression analysis then can serve at the benchmark for providing objective guidance in a variety of settings.
All of this brings me back to where I started and the importance of avoiding HiPPO’s. The HiPPO or Highest Paid Person’s Opinion is how many crucial business decisions are made today. Yet that model is flawed because it creates subjectivity and key man risks related to crucial business decisions. In an uncertain world, adopting objective data-driven methods can help to create consistency for businesses and reduce risk in the process.
Mike McDonald is a PhD in finance and a university professor in the subject at Fairfield University in Connecticut. He also runs a consulting company doing work on quantitative investing, big data, and machine learning for a variety of financial firms, asset managers, institutional investors, and government regulators. Prior to getting his PhD, Mike worked for a major Wall Street bank and one of the top hedge funds. Comments, questions, and concerns are always welcome – email Mike at M.McDonald@MorningInvestmentsCT.com or visit his firm’s website at www.MorningInvestmentsCT.com