Friday, June 8, 2012

Decision Making Under Uncertainty


How do you make important decisions when you're not sure what all the choices are, nor what the effect would be of all the choices, nor even which outcome you most want? I spend a lot of time thinking about this problem, both in general and for specific instances (climate change, the economy, social interactions with friends, etc.). This post made a few interesting suggestions that I see as broadly applicable:

1. Small, incremental steps have an advantage over big policy swings because they are "corrigible," i.e. reversible. From this perspective, timidity can be a good thing, e.g. in a Congress[1], because the small initial steps help build consensus once they start taking effect.

" A wise policy-maker consequently expects that his policies will achieve only part of what he hopes and at the same time will produce unanticipated consequences he would have preferred to avoid. If he proceeds through a succession of incremental changes, he avoids serious lasting mistakes in several ways."

2. Policy choices function as revealed preferences, even if you can't articulate precisely what your desires are. Focus on the policy and not the utility analysis.

"The test is agreement on policy itself, which remains possible even when agreement on values is not."

For instance, I used to get frustrated when I couldn't figure out why a girl agreed to a particular date, e.g. sushi lunch Wednesday. Is she just being nice, or does she just like sushi, or does she think we might be interested in each other? This principle says, "It doesn't matter. Do you want to eat sushi with her? Then go eat sushi."

3. It may be valuable to delay decisions until the situation changes.

"Consider options themselves as assets. Try to retain them or create new ones."
"Seek alternatives whose consequences are observable."
"Plan to allocate resources for monitoring consequences and (if appropriate) gathering more information."

Fencing analogy: I usually do better when I remember to be patient. Eventually the other guy makes a mistake and I can exploit it, but that doesn't happen when I rush to create my own openings, or commit to an all-out attack just because I can't think of anything better to do.

4. Build in resilience.

"Seek alternatives that are 'robust' regardless of outcome." 

This is a weaker than a "dominant strategy" in game theory: the decision doesn't have to be better than every other decision under every possible set of circumstances, but it has to be pretty good regardless of circumstances. Of course, dominant strategies are always robust if you can find them.

5. Be patient.

"Don't assume that getting rid of ignorance and uncertainty is always a good idea."
"Build and utilize relationships based on trust instead of contracts... Trust relationships are more flexible and robust under uncertainty."

Uncertainty is uncomfortable, but it may be better to tolerate uncertainty than to pay the costs of getting rid of it. It depends upon what the cost is. See point #3 above for an example. "DTRs" between friends could be another example, if it puts stress on the friendship.

All of these things except #4 are 180 degrees opposite to my instincts, which makes them interesting to think about. Any thoughts/recommendations?


[1]  I could use a bit more timidity myself. I tend to charge ahead recklessly. Did I ever tell you about my former plan to inject myself with HIV to test whether it really causes AIDS, per Koch's postulates? 

Hahahahaaaa!!! That is ME laughing at YOU, cruel world.
    -Jordan Rixon

I could not love thee, dear, so much,
Loved I not Honour more.

No comments: