Breaking down optimal decision making

In my previous post, I described a possible decision making scenario, and how things might progress. However, there are a lot of motivations I left out. Below are some of my main principles when trying to decide optimally.

• Make sure you are asking the right question. This is the main premise of the blog; If you don't ask the right question, you are unlikely to get the right answer. In the gym example, I was choosing the best gym for me. The process would have been completely wrong if I was picking a place to work out with a friend or to make a recommendation to my parents.
• Figure out the range of possible outcomes. If all of the choices lead to the same outcome, why waste any time deciding between them? If the outcomes are on a wide spectrum from terrible to fantastic, it will be worthwhile to spend some time eliminating possibilities (or defining criteria for an acceptable choice. More on that later).
• Keep in mind what "approximately the best" means.  If you are making a decision with a lot of uncertainty, trying to find the absolute best choice is likely to be a waste of time since you don't know how the options truly compare.
• Target low-hanging fruit first. If you start with a set of possibilities, the order you eliminate options matters (at least when you are trying to decide optimally). In the gym example, we put off the time-consuming task of calling each place to find out if they had a pool until after we had made a short list of those within a two mile radius of us.
• Know when to stop. Once you have a set of options that are approximately the best, it is no longer worth it to spend a lot of time making your final selection. I will sometimes pick an arbitrary criteria to justify whichever choice I want to make at that point (they were the only gym with pretty paintings in the changing rooms).
• You don't have to consider every option. "Satisficing" means you choose the first option you come across that satisfies a set of requirements. This technique can be used at any point in the decision-making process. In the gym example, you might visit gyms that passed the distance and pool criteria, and pick the first one you liked. 

Those are the main tricks I use, whether I am deciding which type of lettuce to buy at the grocery store, or which graduate school to go to. The last tip I have is to try to develop a process you can be happy with ex-poste (after the fact). Occasionally I participate in economics experiments. While I might actually be somewhat risk averse, I usually choose to maximize my expected value since that is objectively the 'best' decision.

Post your tips in the comments!

Are you making the optimal decision, or are you deciding optimally?

Every day you are faced with making decisions. You don't always have all the information, and a big part of deciding optimally refers to the trade-off between spending time gathering information and making the decision.

Since deciding optimally means you have to make decisions without all the information, how can you possible know that you have enough? You might initially think that the amount of data you need depends on how important the decision is. But what if the best option is clear? This suggests that what we really care about is the range of possible outcomes.

Rather than detailing all the things you might consider, I'll describe a possible decision-making process. Say you move to a new town and want to pick out a gym. The list of available options is sitting in your phone book. But are all of those actually options? I like swimming, so any gym I choose will need to have a pool. Based on my experience, a lot of gyms have pools, so I would probably eliminate any far away gyms first, and then find out which of those left have pools. Lastly, I would want to rule out any terrible options by looking up reviews. At this point, any of the options I have left meets all my needs, so the difference between the worst and the best is not so large. I could still spend all week trying to make a decision, or I could just pick a criteria, and go with it. In my case I would just go with whichever option is cheapest, but you could pick the cleverest name or the prettiest website too. Without too much time, I've made a decision that should be, approximately, the best. 

In short, to decide optimally you should consider your options, eliminate options based on your priorities until all remaining possibilities seem to be approximately the same goodness, and then pick any easy to know criteria to choose based on. This won't ensure you always make the best decisions, but it does give you a choice you can stand behind.

What is type three error?

When using data to answer a question, your results might not be perfect. Lets think about two scenarios: 

• Lets say you are trying to figure out if heads and tails are equally likely when you flip a coin. You flip a coin 6 times, and get THTTTH. That's odd... they should be the same. But numbers don't lie right? So you decide that your original hypothesis that they were equally likely is wrong.
• Having lost all confidence in your understanding of probability, you set out to test something you know is different, your expected winnings vs. expected losses when gambling. So you buy $100 worth of lottery tickets, and surprisingly receive $100 in winnings. You conclude that you are expected to break even in the long run.

In statistics, the first case is called "Type 1 error," deciding that the hypothesis that there is no difference (called the null hypothesis) is false when it is true. The second case is called "Type 2 error," deciding that there is no difference, when there is one. 

Understanding that statistics can be misleading is important when making choices. But even more important is asking the right questions. The focus of this blog is avoiding this "Type 3 error," where you might get the right solution to a problem, but it is not really the problem you wanted to solve. The really unique thing about Type 3 error is that it can be eliminated completely by making sure you know which question you need to answer.

First Post!

We are not always certain of the facts when we are trying to make decisions. To quote a famous politician: "There are known knowns, known unknowns, and unknown unknowns." In this blog I'll be exploring decision making and trying to explain rational ways to account for both the knowns, and the unknowns. 

This area is both personally interesting to me as I make decisions about food, transportation, health, and many aspects of life. But it is also academically interesting to me as a PhD student in Industrial Engineering. In my professional life, we call this approach "Optimization under uncertainty." But you don't need a fancy name to use the basic concepts in your daily life. I'm hoping that over the course of my posts, I'll be able to convey my general philosophy of making 'good decisions,' and what that means to me.