• Investing Insights & Exclusive Offers → Get Our FREE “Market Brief”
    Sign-up for our free weekly newsletter. Get unparalleled investing insights and exclusive Summer Sale discounts on Hedgeye research.

    Disclaimer: By joining our email marketing list you agree to receive marketing emails from Hedgeye. You may unsubscribe at any time by clicking the unsubscribe link in one of the emails. Use of Hedgeye and any other products available through hedgeye.com are subject to our Terms Of Service and Privacy Policy

This guest commentary was written by Mike Shell. It was originally published on ASYMMETRY® Observations on 6/14/20. 

A Quantitative Technical Analysis Of COVID-19 - 02.26.2020 black swan cartoon

In addition to the equity markets entering a higher risk level of a drawdown and volatility expansion, we now have a renewed risk of the scary COVID narrative driving more fear.

People tend to overreact and under-react to new information.

We observe it in the global capital markets more frequently than anywhere, and with immediate feedback.

So, those of us who are adept at identifying and monitoring directional trends in global markets have an advantage in researching trends of all kinds if our quantitative trend methods are robust, and our qualitative judgment and decision-making process is repeatable.

Intellectual skills that are associated with acquiring reliable information about nature are parts of the scientific process. Scientists of all kinds need skills like: communicating, observing, classifying, measuring, predicting, inferring, and researching.

Science isn’t just science, there’s also some art to it. I believe the first skills are more art, such as communicating, observing, classifying.

Many say investing and investment management is both art and a science. Some believe investment management is more art than science, others believe it’s more science than art. Quants try to make it more scientific than artistic.

I do a combination. I am Man + Machine.

The way I look at trends and how time series interact with each other is a robust process that may be applied to anything.

I don’t read articles in Bloomberg or The Wall Street Journal to hear the opinions of others to decide what I believe for myself. When I was a young rookie I did read a lot at first, as we all do, then learned the hard way to focus my efforts on my own original research and thinking.

I do best when I do my own work, as an independent thinker.

So, over the decades as a professional researcher, I first inspect the data to observe any trends and then make sense of it afterward. Sometimes my intention is for predictive analytics, other times it’s just prescriptive. predictive and prescriptive analytics. Predictive analytics provides us with the raw information for making informed decisions, while prescriptive analytics provides us with data-backed, evidence-based decision points that we can weigh against one another.

The scientific method is an empirical method of acquiring knowledge through careful observation, and applying rigorous skepticism about what is observed, realizing how cognitive assumptions and bias can distort how one interprets the observation.

  • Descriptive Analytics, applying data aggregation and data mining to provide observations and insight into the past to answer:
    • “What has happened?”
  • Predictive Analytics, applying statistical models and probabilistic forecasting methods to understand the future and answer:
    • “What could happen?”
  • Prescriptive Analytics, applying simulation and testing algorithms to advise on possible outcomes and answer:
    • “What should we do?”

The Scientific Method and experimenting is a systematic approach to problem-solving and decision-making.

An algorithm may look something like this:

Problem → Hypothesis → Prediction → Test Predictions → Evaluation

We all have biases. All industries have biases. Sometimes these biases gave blindspots. Our biases that can narrow our vision and influence behavior and beliefs. It’s why in asset management, we often consult with researchers outside the industry to help avoid blind spots from industry bias. For example, the personal financial planning profession has a tendency to blindly say “balance your risks and rewards” and “balance your portfolios”, which is about the silliest things I’ve ever heard.

If you balance your risk and reward, you get symmetry on your statement.

If we want asymmetry, we have to skew the risk and reward positively.

It’s essential to identify blind spots, own them without being defensive, and adjust our behavior to avoid it.

MY OBSERVATIONS OF COVID 19

I have an advantage, because I observe COVID 19 trends and rates of change as it is, without any bias as to beliefs about the disease and such. That is, I’m just purely looking at the data we collect and feed into our systems for observation.

It’s like this:

  • I focus on; what has changed?
  • I look for extremes in levels, like new high or low breakouts.
  • I also monitor the rates of change. Fast breakouts are more likely to form an ongoing trend than slow.

FLORIDA COVID 19 UPDATE

As COVID 19 and the mass quarantine strategy for suppressing the spread has become a political debate lately, so some of you may perceive what you are about to read that way.

Don’t.

I have no political bias about this whatsoever. My personal preference is to get past this virus as quickly as possible with as little human suffering as possible. Ignoring the data and facts doesn’t get us there. Exaggerating the data and facts doesn’t either. So, I suggest you try to see the trends for what they are, as I am.

My home state of Florida is now, unfortunately, trending in the wrong direction. I pointed it out with some fellow money manager friends last week of a potential breakout in the trend and it has since trended higher for a meaningful and material breakout.

No one wants businesses to open and get back to normal more than me, but what is, is. The cases per day is in a strong uptrend. The prior high was 1575 on April 3rd and 1601 on May 16th. Yesterday was 2581. I hope to see this trend down.

A Quantitative Technical Analysis Of COVID-19 - 6 15 2020 10 39 45 AM

NEW UPTREND IN FLORIDA CASES NOT DRIVEN BY NEW TESTING

The first hypothesis we think of is, well, maybe the uptrend in driven by an increase in testing. Naturally, increasing the absolute testing also should increase the number of positive tests. That isn’t the case. The high in testing was May 20th in Florida. The testing per day remains materially below that level according to the most recent data.

A Quantitative Technical Analysis Of COVID-19 - 6 15 2020 10 42 45 AM

Keep in mind, the lower line is tests administered per day, so there is a lag between testing and the classification of a positive case. In fact, there are natural lags in all of this data.

For example, I don’t expect to see the results of the protests until a week or two afterward to account for the lag in showing symptoms, going to get tested, and getting the test results. If there is any increase in the protesting areas, we’ll hear about it next week or later.

FLORIDA COVID 19 HOSPITALIZATIONS AT AN ALL TIME HIGH

Not much to add here. It is what it is.

A Quantitative Technical Analysis Of COVID-19 - 6 15 2020 10 45 00 AM

I can try to make it seem better with a logarithmic based chart, which draws the chart in a way that two equal percent changes are plotted as the same vertical distance on the scale. It visually normalizes the rate of change.

The good news is the rate of change overall is slowing. The bad news is this could look like an S-curve later, which would be typical of a spread.

A Quantitative Technical Analysis Of COVID-19 - 6 15 2020 10 46 07 AM

By the way, here is the log chart of the new cases per day. We normally use a logarithmic chart scale for long term charts to normalize the data especially if I’m comparing it to something else where relative strength (rate of change) is measured. But here, we still see a breakout in rate of change. So, it’s a material breakout in my opinion, but I hope it breaks down.

A Quantitative Technical Analysis Of COVID-19 - 6 15 2020 10 48 12 AM

Unfortunately, three of the new uptrends are in states were we have clients; Florida, California, and Texas. Next up is Texas.

TEXAS COVID 19 UPTREND

New cases in Texas is trending up to all time highs. I think Houston, Texas is now at risk of another stay-at-home order.

A Quantitative Technical Analysis Of COVID-19 - 6 15 2020 10 53 22 AM

CALIFORNIA COVID 19 TREND

California new cases per day has trended up to the all time high again. It doesn’t seem to be in direct result of more testing per day, either.

A Quantitative Technical Analysis Of COVID-19 - 6 15 2020 10 54 19 AM

ARIZONA COVID 19 CASES AT NEW HIGH

Arizona is at a high in new daily cases reported, but also in tests per day. The trouble in Arizona is the material new uptrend in hospitalizations.

A Quantitative Technical Analysis Of COVID-19 - 6 15 2020 10 55 16 AM

The momentum in hospitalizations in Arizona is a real problem, and I’m using a logarithmic scale below, but it doesn’t help.

A Quantitative Technical Analysis Of COVID-19 - 6 15 2020 10 58 41 AM

So, we are seeing new hospitalization highs in Texas, Arizona, and North Carolina. We’re showing Louisiana here as well for context, since it had an early COVID-19 outbreak.

A Quantitative Technical Analysis Of COVID-19 - 6 15 2020 11 00 14 AM

We saw 3 states report more than 2,000 cases yesterday: California, Florida, and Texas. A picture speaks a thousand words.

A Quantitative Technical Analysis Of COVID-19 - 6 15 2020 11 08 35 AM

A Quantitative Technical Analysis Of COVID-19 - 6 15 2020 11 10 40 AM

But the South is now seeing a surge in cases…

A Quantitative Technical Analysis Of COVID-19 - 6 15 2020 11 11 51 AM

And it doesn’t seem to be an increase in testing.

A Quantitative Technical Analysis Of COVID-19 - 6 15 2020 11 12 36 AM

Can hospitalizations keep falling if cases are rising?

A Quantitative Technical Analysis Of COVID-19 - 6 15 2020 11 14 13 AM

Because hospitalizations are rising rapidly in some Southern states.

A Quantitative Technical Analysis Of COVID-19 - 6 15 2020 11 14 13 AM

I expect we’ll be hearing a lot about these new high breakouts in Arizona, Florida, Texas, and California. All of which, by the way, are the hottest and most humid states in the U.S., so much for the heat and humidity killing the virus.

What we have here is, a a notable uptrend across the South. I hope to see it fade, but based on what I’m seeing, it’s more likely to continue. Only time will tell.

If you are at risk, I recommend remaining cautious, wearing the dang mask, and treating this virus with respect.

ABOUT MIKE SHELL 

Mike Shell is the founder of Shell Capital Management, LLC and the portfolio manager of ASYMMETRY® Global Tactical and ASYMMETRY® Managed Portfolios.

This piece does not necessarily reflect the opinion of Hedgeye.