Don’t Expect the Expected Value

One day, over 50 years ago, we were visiting Lake Tahoe as a family, and dad went across the border to play keno. He came back elated: he had hit seven out of eight on one of his tickets, and won eleven hundred dollars. He proudly laid out fifty twenties and two fifties on the kitchen table. It was a magnificent sight.

The details of keno are unimportant here, except to note that keno is not a game of skill. Of course the house has an edge. In the long run, you will lose money playing keno no matter how you do it. Even my dad, who over the years has played a lot of keno, and won even bigger payouts, would probably admit that he might have a net lifetime loss.

So why do people play? There are lots of reasons, I’m sure, but one of them must be connected to that heartwarming anecdote: fifty years later, I remember the event clearly, as one of joy and wonder.

Let’s explore that using roulette, which is much simpler than keno. A roulette wheel has 18 red and 18 black numbered slots, plus a smaller number of green slots (often two). You can make many different bets, but we will stick with red and black. If you place a $1 bet on red, and it comes up red, you get $2 back (winning $1); if it comes up black or green, you lose your dollar.

Continue reading Don’t Expect the Expected Value


Trees. And. Diagnosis. (Part two)

Last time we introduced decision trees and a tool we’ve made to explore them. With that tool, embedded in a simple game (Arbor), you can generate data from alien creatures with a simulated malady, figure out its predictors, and make a decision tree that will let you automate its diagnosis. (Here is the link to that not-quite-game.)

Your job was to get through the diseases ague and botulosis. Today I want to reflect on those two scenarios.


Ague is ridiculously simple, and with that ridiculous simplicity, the user is supposed to be able to learn the basics of the game, that is, how to “drive” the tools. One way to figure out the disease is to sort the table by health and see what matches health. Here is what the sorted table looks like:


Just scanning the various columns, you can see that health is associated with hair color.  Pink means sick, blue means well. With that insight, you can go on to diagnose individual creatures and then make a simple tree, which looks like this:


Although there is a lot of information in the tree, users can generally figure it out. If they (or you) have trouble, they can get additional information by hovering over the boxes or the links.

Continue reading Trees. And. Diagnosis. (Part two)

Trees. And. Diagnosis. (Part one)

(This is part one. Link to part two.)

In the Data Science Games project, we have recently been exploring decision trees. It’s been great fun, and it’s time to post about it so you dear readers (all three or so of you) can play as well. There is even a working online not-quite-game you can play, and its URL will probably endure even as the software gets upgraded, so in a year it might even still work.

Here’s the genesis of all this: my German colleague Laura Martignon has been doing research on trees and learning, related to work by Gerd Gigerenzer at the Harding Center for Risk Literacy. A typical context is that of a doctor making a diagnosis. The doctor asks a series of questions; each question gets a binary, yes-no answer, which leads either to a diagnosis or a further question. The diagnosis could be either positive (the doc thinks you have the disease) or negative (the doc thinks you don’t).

The risk comes in because the doctor might be wrong. The diagnosis could be a false positive or a false negative. Furthermore, these two forms of failure are generally not equivalent.

Anyway, you can represent the sequence of questions as a decision tree, a kind of flowchart to follow as you diagnose a patient. And it’s a special kind of tree: all branchings are binary—there are always two choices—and all of the ends—the leaves, the “terminal nodes”—are one of two types: positive or negative.

The task is to design the tree. There are fancy ways (such as CART and Random Forest) to do this automatically using machine learning techniques. These techniques use a “training set”—a collection of cases where you know the correct diagnosis—to produce the tree according to some optimization criteria (such as how bad false positives and false negatives are relative to one another).  So it’s a data science thing.

But in data science education, a question arises: what if you don’t really understand what a tree is? How can you learn?

That’s where our game comes in. It lets you build trees by hand, starting with simple situations. Your trees will not in general be optimal, but that’s not the point. You get to mess around with the tree and see how well it works on the training set, using whatever criteria you like to judge the tree. Then, in the game, you can let the tree diagnose a fresh set of cases and see how it does.

That’s enough for now. Your job is to play around with the tool. It will look like this to start:


The first few scenarios are designed so that it’s possible to make perfect diagnoses. No false positives, no false negatives. So it’s all about logic, and not about risk or statistics. But even that much is really interesting. As you mess around, think about the representation, and how amazingly hard it can be to think about what’s going on.

There are instructions on the left in the tan-colored “tile” labeled ArborWorkshop. Start with those. There is also a help panel in the tree tile on the right. It may not be up to date. All of the software is under development.

Here is the link:

The first disease scenario, ague, is very simple. The next one, botulosis, is almost as simple, and worth reflecting on. That will happen soon, I hope after you have tried it.

(Part two. About Ague and Botulosis.)

Note: if you are unfamiliar with this platform, CODAP, go to the link, then to the “hamburger” menu. Upper left. Choose New. Then Open Document or Browse Examples. Then Getting Started with CODAP. That should be enough for now.

The Index of Clumpiness, Part Three: One Dimension

In the last two posts, we talked about clumpiness in two-dimensional “star fields.”

  • In the first, we discussed the problem in general and used a measure of clumpiness created by taking the mean of the distances from the stars to their nearest neighbors. The smaller this number, the clumpier the field.
  • In the second, we divided the field up into bins (“cells”) and found the variance of the counts in the bins. The larger this number, the clumpier the field.

Both of these schemes worked, but the second seemed to work a little better, at least the way we had it set up.

We also saw that this was pretty complicated, and we didn’t even touch the details of how to compute these numbers. So this time we’ll look at a version of the same problem that’s easier to wrap our heads around, by reducing its dimension from 2 to 1.  This is often a good strategy for making things more understandable.

Where do we see one-dimensional clumpiness? Here’s an example:

One day, a few years ago, I had some time to kill at George Bush Intercontinental, IAH, the big Houston airport. If you’ve been to big airports, you know that the geometry of how to fit airplanes next to buildings often creates vast, sprawling concourses. In one part of IAH (I think in Terminal C) there’s a long, wide corridor connecting the rest of the airport to a hub with a slew of gates. But this corridor, many yards long, had no gates, no restaurants, no shoe-shine stands, no rest rooms. It was just a corridor. But it did have seats along the side, so I sat down to rest and people-watch.

Continue reading The Index of Clumpiness, Part Three: One Dimension

The Index of Clumpiness, Part Two

Last time, we discussed random and not-so-random star fields, and saw how we could use the mean of the minimum distances between stars as a measure of clumpiness. The smaller the mean minimum distance, the more clumpy.

Star fields of different clumpiness, from K = 0.0 (no stars are in the clump; they’re all random) to K = 0.5 to K = 1.0 (all stars are in the big clump)

What other measures could we use?

It turns out that the Professionals have some. I bet there are a lot of them, but the one I dimly remembered from my undergraduate days was the “index of clumpiness,” made popular—at least among astronomy students—by Neyman (that Neyman), Scott, and Shane in the mid-50s. They were studying Shane (& Wirtanen)’s catalog of galaxies and studying the galaxies’ clustering. We are simply asking, is there clustering? They went much further, and asked, how much clustering is there, and what are its characteristics?

They are the Big Dogs in this park, so we will take lessons from them. They began with a lovely idea: instead of looking at the galaxies (or stars) as individuals, divide up the sky into smaller regions, and count how many fall in each region.

Continue reading The Index of Clumpiness, Part Two

Capture/Recapture Part Two

Trying to get yesterday’s post out quickly, I touched only lightly on how to set up the various simulations. So consider them exercises for the intermediate-level simulation maker. I find it interesting how, right after a semester of teaching this stuff, I still have to stop and think how it needs to work. What am I varying? What distribution am I looking at? What does it represent?

Seeing how the two approaches fit together, yet are so different, helps illuminate why confidence intervals can be so tricky.

Anyway, I promised a Very Compelling Real-Life Application of This Technique. I had thought about talking to fisheries people, but even though capture/recapture somehow is nearly always introduced in a fish context, of course it doesn’t have to be. Here we go:

Human Rights and Capture/Recapture

I’ve just recently been introduced to an outfit called the Human Rights Data Analysis Group. Can’t beat them for statistics that matter, and I really have to say, a lot of the explanations and writing on their site is excellent. If you’re looking for Post-AP ideas, as well as caveats about data for everyone, this is a great place to go.

One of the things they do is try to figure out how many people get killed in various trouble areas and in particular events. You get one estimate from some left-leaning NGO. You get another from the Catholics. Information is hard to get, and lists of the dead are incomplete. So it’s not surprising that different groups get different estimates. Whom do you believe?

Continue reading Capture/Recapture Part Two