Home » Blog » Insights » Complexity Modeling – Pandemic and More

Complexity Modeling – Pandemic and More

complexity modeling

Complexity modeling is something we humans do in abundance. Let’s dig into why we model and what it means.

We hear the word “model” a lot these days, first about climate and now about COVID-19. Simply put, biological systems anticipate their environment. Humans use our brains to build anticipatory models to guide our actions. For really tough problems, we use mathematics to extend our mental models.

You have probably heard that “all models are wrong, but some are useful”. Models are used to explore relationships and changes that may occur. We create formulas using parameters and coefficients. Parameters describe perceived relationships and coefficients describe change in output versus change in parameter.

Epidemic analysis has been around for a long time. We use formulas to estimate spread of a disease, or its effective reproduction. Angela Merkel does a good job of explaining these models. Basically, we want to get the effective reproduction rate below 1.

You can see above the wide range of complexity modeling results for New York City provided by various experts, compared to the actual numbers in Purple as of early April. These models were all “wrong but useful” in helping NY prepare for worst case scenarios. It appears that the models underestimated the positive effects of mitigation (distancing, etc.) enacted by governors/premiers and local authorities.

(Keep in mind the paucity of good data, despite all the numbers you see on TV. The only real data is institutional, i.e. from hospitals and nursing homes. All other data is a somewhat informed wild ass guess. No one really knows real infection or antibody rates. Plus, there are so many different organizations doing different modeling, and countries collecting data in different ways. Sigh. Kudos to the experts for doing their best.) Check out this cartoon about pandemic modeling.

Complexity Modeling – Climate Uncertainties

Complexity is uncertainty about order and explanation. Complexity modeling of epidemic spread has helped some leaders understand and reduce uncertainty during these difficult times. In particular, scaling our health resources to meet population need.

I am sure that the environmental activists are ramping up to use COVID-19 as a warning about the future effects of climate change. What I find interesting is that so-called climate models are orders of magnitude more complex than epidemic models. Not even close. Climate models contain many more layers of abstraction, theory and non-linearity than models for COVID spread.

What’s more, we have been able to assess the usefulness of the pandemic models over a relatively short period of time (weeks and months.) Our ability to build and evaluate climate models has been spread over centuries for proxy collection and will spread over decades for evaluating efficacy.

We can clearly see that our pandemic interventions (social distancing, testing, tracing, etc.) have worked to a large degree. As for climate, we may never know if shutting down Alberta’s energy sector makes any difference, if indeed there is a difference to be made.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.