A Fox – Like Approach to Forecasting
From Nate Silver’s book The Signal and the Noise
How Foxes think
Multidisciplinary: Incorporate ideas
from different disciplines and regard-
less of their origin on the political
spectrum.
Adaptable: Find a new approach
or pursue multiple approaches at the
same time-if they aren't sure the
original one is working.
Self-critical: Sometimes willing (if
rarely happy) to acknowledge mistakes
in their predictions and accept the
blame for them.
Tolerant of complexity: See the uni-
verse as complicated, perhaps to the
point of many fundamental problems
being irresolvable or inherently unpre-
dictable.
Cautious: Express their predictions in
probabilistic terms and qualify their
opinions.
Empirical: Rely more on observation
than theory.
Foxes are better forecasters.
For forecasting principals:
Think Probabilistically
Today’s forecast is the first forecast of the rest of your like (ie, be willing to update forecasts as more information is known and noise is separated from the system). Aka, think marginally, don’t be the victim of the endowment effect. (I.e., assigning ownership value to the current forecast)
Look for consensus – there is evidence that group forecasts are more accurate than individual ones. Though, the group doesn’t alleviate the responsibility to think critically. Eg, the consensus economic forecast in 2006 was the housing market was strong and would continue to be stable well into the future….
The 2nd principal reminds me of Fratto’s thoughts on adaptability.
