The Desperate Search For Peak Coronavirus

The most important economic indicator at the moment has nothing to do with the dismal science per se and everything to do with how a certain health crisis unfolds over the days, weeks and months ahead. The elephant in the room, of course, is coronavirus data. In particular, the point where it peaks. Depending on the country, the timing (and reliability) of the peaking varies. Ultimately, a global apex is the main event. Unfortunately, the data is questionable and the future is unclear as ever. Not a great basis for analysis and forecasting, but it’s the only game in town. With that in mind, let’s take a brief tour of US data to see where we stand.

In this quick overview, I’ll focus on reported deaths, which are arguably more reliable in some degree vs. reported cases of infections. The numbers are published by Johns Hopkins. The first chart below shows how the daily change in daily fatalities has evolved (through yesterday, Apr. 8). For the moment, there’s no sign of peaking. Earlier this week there was a hint that an apex in daily deaths for the US might be near, but the updates didn’t play along and so we’re still in a period of an increasing trend. A simple forecasting model (based on combining models) suggests that the reported deaths will continue to rise over the next several days. (Warning: the forecasting methodology is solely based on the data, rather than a robust epidemiological model, and so the obvious caveats apply.)

Tracking the deaths by the cumulative total looks even worse.

Presenting the cumulative numbers in log scale provides what is arguably a more realistic view of the trend by adjusting for the dramatic changes in recent vs. older results. By this standard, a peak may be approaching… if you squint hard enough.

Another way to view the data: daily percentage changes. On this front, there’s modest support for arguing that the trend has already peaked and a slow fade is underway. For deeper perspective we might look to moving averages over, say, 5- and 10-day windows.

The good news is that a peak is coming… eventually—for the US and the world. Exactly when that peak arrives, unfortunately, remains a mystery. A more sophisticated forecasting model (developed by the University of Washington’s Institute for Health Metrics) is currently projecting that US deaths will peak in a few days – Apr. 12, based on the median estimate. But as always with modeling the future, there’s a wide band of uncertainty to consider.

Alas, with so many unknowns, extreme caution is still required for estimating a future peak. For example, health experts warn that Covid-19 may not fade in the summer.

Learn To Use R For Portfolio Analysis
Quantitative Investment Portfolio Analytics In R:
An Introduction To R For Modeling Portfolio Risk and Return

By James Picerno

But there are hints of progress to consider, too. That includes signs that New York – the hardest hit region in the US – appears to be flattening the coronavirus curve.

By other measures, however, “things are still getting worse.” As CNN reports today, “The US death toll crossed 14,000 on Wednesday, with a record 1,858 deaths reported just on Tuesday. Since the outbreak started, about 425,000 cases have been diagnosed in the US. And researchers say the peak has yet to come.”

How is recession risk evolving? Monitor the outlook with a subscription to:
The US Business Cycle Risk Report

One thought on “The Desperate Search For Peak Coronavirus

  1. Pingback: How Health Crisis Unfolds Has Become an Important Economic Indicator -

Comments are closed.