Discussion about this post

User's avatar
PenguinEmpireReports's avatar

Yes, you’re correct that models are merely assumptions fed into a computer. Having built relatively simple forecasting models for the agricultural business, it’s humbling to see how a slight changes can have big impacts. We would be proud if the model would be off by 1% of the forecasted values, in a 1 year timeframe. But over 5-10 years, that 1%/ per year compounds quickly and the next thing you know, you can be way off. Especially if you misread when and where the business cycles change. And that’s with a good model based on some of the most accurate data available that is carefully measured.

And that’s nothing compared to the vast complexity of our climate where we have difficulty measuring impacts in the real world.

Data ‘slippage’ from the real world, missed cycle cycles, and the margin of error can easily compound until the model is not more use than a pile of garbage. Especially if you don’t check assumptions.

Expand full comment
Thomas J Shepstone's avatar

Excellent! Peter Ridd was on Alexandra Marshall today talking about coral:

https://watch.adh.tv/alexandra-marshall/season:1/videos/peter-ridd-alexander-voltz-monday-18-december-2023

Expand full comment
2 more comments...

No posts