4 ms·
A model isn't a crystal ball, it is a way of formally describing something using mathematics. A model could tell you that a specific set of conditions that prec
by jklein11 8y ago
A model isn't a crystal ball, it is a way of formally describing something using mathematics. A model could tell you that a specific set of conditions that preceded a recession have occurred again. One example of this would be an inverted yield curve.[1]
A recession is when an economy declines for at least six months.[2] I guess the best predictor of a recession is 5.5 months of negative growth.
1. https://www.bloomberg.com/opinion/articles/2018-12-03/u-s-yield-curve-just-inverted-that-s-huge https://www.bloomberg.com/opinion/articles/2018-12-03/u-s-yi...
2. https://www.thebalance.com/what-is-a-recession-3306019 https://www.thebalance.com/what-is-a-recession-3306019
- amelius 8y agoYou probably missed the word "obvious" in my comment. I was just saying that if economists don't predict a recession, then the chances are pretty low that parent-poster could successfully predict one based on intuition only.