Author Archives: admin

An Ode to Subjective Power Rankings

Leave a reply

power rankings

Any good standings page has a wealth of objective team information. Season wins/losses. Points scored for/against. Record in the last 10 games. Games behind first place. Etc.

What purpose do “Power Rankings” serve on top of objective standings? Some experts watched the sport and decided that a team was perhaps outperforming its record? They had some witty comment to make for a few teams and decided to go ahead and try to right something down for all the teams?

Experts aren’t very accurate in their predictions. Power rankings are no exception to this rule. I choose not to read power rankings any more. The very occasional witty comment is overwhelmed by the mass of obvious/rehashed observations that make up the power rankings commentary. This is part of a general trend in me listening to ESPN less and less and looking at actual performance data more and more.

Theory Tuesday- Prospect Theory

Leave a reply

Prospect Theory, created by Kahneman and Tversky, is a way to quantify utility functions for people. In traditional economics, a dollar is a dollar and people are expected to maximize their wealth. In reality, people have biases. The two major biases shown in prospect theory are diminishing sensitivity and loss aversion.

Diminishing sensitivity makes people notice $1 changes in their wealth less the farther that they get from their reference point. $1 means a lot more to someone who is close to breaking even than to someone else who has already made a million dollars.

Loss aversion refers to the fact that people feel losses more strongly than equal sized gains. A loss of $100, relative to your reference point, feels worse than a gain of $100 feels good.

The effects can be seen in the following pictures. (a) isolates diminishing sensitivity and (b) isolates loss aversion. Their combined effects are shown in (c).

RplotAllThree

Two Links Tuesday- June 3, 2014

Leave a reply

ACM Web Science 2014 Conference at IU. This conference is coming to Bloomington June 23-26. It is a single-track conference that is a little far afield from what I usually study, but I plan to try to sneak in nonetheless. If you want to come with me and cause a diversion while I sneak in, that would be appreciated.

How far your paycheck goes in various cities. How is Bloomington the poorest city (via median income) in the country? That doesn’t seem right. Are they counting in all the part-time students income? I think Bloomington is about 40k students and about 40k permanent residents, so the median might get significantly pushed down by including all the part-time students.

Life Tips- Bouncy Ball Chair

Leave a reply

Your desk job is slowly killing you. Sitting all day in a chair is bad. Bad for posture, bad for health. While there are many possible solutions, I’m trying out sitting on a 75” stability exercise ball (bouncy ball for short) this summer.

bouncy ball

At first, it was uncomfortable and made my lower back tired with all the balancing. A week in, however, I don’t feel any pain and really enjoy the alternative chair. My hope is that the balancing and bouncing eliminate some of the effects of sitting all day. Even if they don’t, it is undeniable that I am more active with a bouncy ball chair. Being active is good. I use a weighted, stable ball that doesn’t roll away when I get up. If all goes well, I will use the chair when I move back to my school office in the fall. Maria uses a bouncy ball chair at her job at IU as well.

Code Monkey Monday- Eclipse and PyDev

2 Replies

If you program in Python, you’ll eventually want to use a development environment. Eclipse is a well-known IDE (integrated development environment) for Java programming. You can co-opt Eclipse to use with Python by using PyDev.

Download Eclipse, saving it to your desktop. I don’t install it to my C: drive or anything; I’m not even sure if you can do that. Download PyDev. Drag the folders “features” and “plugins” to your Eclipse folder, thus merging the contents of the PyDev folders and the Eclipse folders. Open Eclipse and select a working directory. In Eclipse, goto Window->Open Perspective and PyDev should be listed there. Open it up and now you’ll be able to program in Python in Eclipse and run Python programs in Eclipse.

Blog Post Schedule

Leave a reply

Here is the current weekly schedule for blog posting for me:
Monday AM: Code Monkey Monday
Monday PM: Life Tips
Tuesday AM: Two Links Tuesday
Tuesday PM: Theory Tuesday
Wednesday: Sports Talk
Thursday AM: Theory Thursday
Thursday PM: Project Updates
Friday: Book Review

What to expect in the posts:
I don’t like to spend a lot of time typing, so most posts will be short. You’re welcome. If you want further elaboration, please comment or email (eric.michael.webb@gmail.com) me.
Code Monkey Monday: I’ve programmed a lot more than most non-computer scientists. I will write tips for getting started and details of interesting things I’ve come across.
Life Tips: Short suggestions of things that make more more healthy/productive/awesome.
Two Links Tuesday: Two websites or news stories that caught my eye and are worth sharing.
Theory Tuesday/Thursday: I’m studying Decision Science and Operations Management with the hope of teaching these subjects in the future. These short posts are my way of providing brief explanations of technical topics for anyone that wants to know what my field is all about. Hopefully future students can reference them.
Project Updates: Short post about what I’ve accomplished lately.
Book Reviews: I read quite a bit, most of which is non-fiction. These are short reviews of the books that I have read recently.

Update on 8/15/14: Occasionally I will write a post about my travels. This will only happen when I have a unique perspective that Maria’s website can’t cover.

A backlog of each kind of post will be available in the “Past Posts” section.

Summer Plans

Leave a reply

The lack of posting lately is directly related to the influx in homework and extracurricular travel that has accompanied this semester. But classes are over in a couple weeks, and I’m planning on creating a regular schedule for posting for the summer. Give me a couple weeks to create a backlog of posts so that the system works.

This summer, I’ve decided not to take on any commitments. No school. No internships. No job. May-August will be the longest streak of unstructured time that I have had in my adult life.

Junior/senior years of high school, I had a summer job. McDonald’s and Lonestar Steakhouse. From 2006-2010, I had 5 summer internships. Hope College, Case Western, UCLA, and 2 years at the Aerospace Corporation. 2011-2013 I was working at Booz Allen. So this year will be the first time I’ve had time to myself.

The goal is to be outrageously productive in research, to compensate for the fact that I’ve had to put research on the back-burner this semester due to homework and travel. Will update as I go. We’ll be in Bloomington for most of the summer, save for a few short trips.

School Schedule

Leave a reply

As a reference, here are the courses I have taken at IU thus far. Recall that I am doing a PhD in Decision Science and Operations Management.

Fall 2013:
Linear Optimization (1/2 semester, Kurt Bretthauer)
Integer and Nonlinear Optimization (1/2 semester, Kurt Bretthauer)
Dynamic Programming (1/2 semester, Goker Aydin)
Exploratory Data Analysis (full semester, Karen Kafadar)
Foundations of Information Systems Research (1/2 semester, Alan Dennis)
Introduction to Complex Systems (full semester, Filippo Radicchi)

Spring 2014:
Inventory Theory (1/2 semester, Gil Souza)
Supply Chain and Distribution (1/2 semester, Gil Souza)
Behavioral Operations Management (1/2 semester, Shanshan Hu)
Humanitarian Logistics (1/2 semester, Alfonso Pedraza Martinez
Game Theory (1/2 semester, Eric Rasmusen)
Asset Pricing Theory (full semester, Craig Holden)

Directional Statistics

Leave a reply

At Booz Allen in 2012, I needed to find the average time of day that something happened as part of my code. Taking the normal (linear) average of times of day doesn’t work because there is always a discontinuity somewhere, usually midnight. Since times of day are periodic, 11:59 pm and 1:01 am should average to midnight, not noon. But they’ll average to noon (1200) if you average something like 2359 (military time) and 0001.

I thought we had stumbled on to an area of mathematics that needed further study: finding the average and variance of periodic variables. Because all of my internet searches were things like “how to average periodic variables” and “average time of day” and the like, I could not find anything on the internet about how to solve this problem. So I started to derive an answer and ended up with something very similar to what I now learn is called Directional Statistics. Wiki article on directional statistics. I never thought to search for anything like “Directional Statistics” or “Circular Statistics” or anything similar, so I didn’t find Directional Statistics until Dave Anderson showed them to me this week.

The basic idea (that AJ Mobley, Dave, and I derived at Booz Allen) is that you treat times of day as points on the unit circle, and then average the points in two-dimensions. Invert the “average point” to get the average time of day. Additionally, the farther the point is from the origin, the lower the variance of the points.