Mind on Money: Data modeling has real-world implications

Mind on Money: Data modeling has real-world implications

Many professional career activities involve a large component of uncontrollable output or results.

A teacher can design the best lesson plans and provide the most engaging educational experience, but she can’t control whether her student got a good night sleep before the standardized test and comes in ready to perform.

An accountant can have the best processes and years of experience, but she can’t control whether her client actually reports their transactions accurately. So it could be said about many professions, and get any two or three of any type of professionals together and the griping about this reality is likely to emerge.

Imagine hanging out with a group of financials advisers (don’t immediately nod off). You would likely hear work-based griping along the lines of “we can design the best plans, based on the smartest methodology and ultimately we can’t control financial markets or how clients will respond to them.”

Despite this insider whining, financial advisers (like most other professions) are expected to be right about directly uncontrollable outcomes (markets and client behavior) more of the time than not.

Like other professions, kids have to perform, books have to balance, and financial plans have to function despite the variables we can’t control. So it also goes with health care policy doctors.

Let’s pretend I’m a health care policy doctor, and my client is the President of the United States (or any state governor). In late March I advise my client that even if drastic measures are taken, U.S. deaths from COVID-19 could reach 100,000 to 200,000 (a ridiculously broad range in its own right) by the end of April. My advice is based on a mathematical model I helped build, and despite the reality that the current actual data is nowhere near the assumptive data I baked into my model, it could be, and I advise with confidence.

I recommend the president or governor use the power of government to put a highly complex $22 trillion economy into what amounts to an induced coma, having no real knowledge or understanding of the long-term unintended outcomes of this recommendation, which could include destroyed businesses, destroyed retirements, destroyed careers and destroyed lives. We have to do this to save lives, we cannot lose 200,000 American lives. We must rely on the models, we must rely on the data.

Then the data continues to emerge. Not only is the real data not consistent with the assumptive data I used in the models I used to advise, it's so outside the range of projected probabilities the model itself becomes almost worthless.

Back in the non-imaginary world, as a financial adviser I understand models. I build and maintain them for my clients every week. I know it would be easy to alter assumptive data in a retirement model, in ways the client wouldn’t even notice, to make the model tell whatever story I want. A subtle adjustment in investment returns, a subtle change in inflation rates and the modeling process can show the same person going broke or becoming a multi-millionaire.

Anyone who builds mathematical models in their work knows this is true. But as experienced professionals we don’t do this. We want to use our experience and the collective knowledge of our profession to develop solid solutions for our clients and employers. In the end, real professionals want their models to be accurate.

Based on the actual data which is released daily in the U.S., it appears impossible at this point that the worst case COVID-19 projections will come even remotely close to being possible. Through Wednesday afternoon of this week the financial markets seemed to be responding to this realization. I for one, am breathing a sigh of relief.

I believe investors will soon begin to grapple primarily with the economic decisions made during the crisis period and on that front, there is still mountains of uncertainty. I pray for the continued wrongness of the experts, and I also pray we don’t face this type of challenge again for a generation at least, because based on the performance of the experts this time, they’ll have a hard time getting us to listen if next time comes to soon.

Opinions are solely the writer's and are for general information only and are not intended to provide specific advice or recommendations for any individual. Stock investing involves risk, including loss of principal. Marc Ruiz is a wealth advisor and partner with Oak Partners and registered representative of LPL Financial. Contact Marc at marc.ruiz@oakpartners.com. Securities offered through LPL Financial, member FINRA/SIPC.