New online course: model thinking
There’s a new course starting soon, taught by Scott Page, about “model thinking” (hat tip David Laxer). The course web site is located here and some preview lectures are here. From the course description:
In this class, I present a starter kit of models: I start with models of tipping points. I move on to cover models explain the wisdom of crowds, models that show why some countries are rich and some are poor, and models that help unpack the strategic decisions of firm and politicians.
The models cover in this class provide a foundation for future social science classes, whether they be in economics, political science, business, or sociology. Mastering this material will give you a huge leg up in advanced courses. They also help you in life.
In other words, this guy is seriously ambitious. Usually around people who are this into modeling I get incredibly suspicious and skeptical, and this is no exception. I’ve watched the first two videos and I’ve come across the following phrases:
- Models make us think better
- Models are better than we are
- Models make us humble
The third one is particularly strange since his evidence that models make us humble seems to come from the Dutch tulip craze, where a linear model of price growth was proven wrong, and the recent housing boom, where people who modeled housing prices as always going up (i.e. most people) were wrong.
I think I would have replaced the above with the following:
- Models can make us come to faster conclusions, which can work as rules of thumb, but beware of when you are misapplying such shortcuts
- Models make us think we are better than we actually are: beware of overconfidence in what is probably a ridiculous oversimplification of what may be a complicated real-world situation
- Models sometimes fail spectacularly, and our overconfidence and misapplication of models helps them do so.
So in other words I’m looking forward to disagreeing with this guy a lot.
He seems really nice, by the way.
I should also mention that in spite of anticipating disagreeing fervently with this guy, I think what Coursera is doing by putting up online courses is totally cool. Check out some of their other offerings here.
mathbabe 
I’m also excited about this class. I do hope that there will be good discussion of how models can fail and mislead, what to do in the face of plausible but competing models, and a message that we should also be humble about models, not just be humbled by them.
LikeLike
Yes, it’s a good idea, and yes, he’s a nice guy, and yes, it’s good that Stanford (and Cloudera) is doing this. But the first segment was almost entirely boiler plate. (See my comments here: http://goo.gl/aGjeV) The second segment was significantly better. (See my comments here: http://goo.gl/DSgBa)
By the way, I doubt that Page would disagree with your comments about models.
LikeLike
i think he also has a complexity course online (not free), i’ve seen a couple of lectures from it and those are really good. I don’t know why he is doing this “anti-complex” lecture series.
LikeLike
I had the same reaction — there’s something about his presentation that suggests this’ll be a “modeling for poets” course. This kind of oversimplification can do harm in the world. sadface.
LikeLike
Any — insider? — idea about when they are going to start? 🙂
Having finished Machine Learning, I’ve signed up for three more, including this one, but they are postponed till future notice…
LikeLike
Don’t know! I’ll try to find out.
LikeLike
The link to the additional courses is Coursera and not Cloudera which is the Hadoop Company:
http://www.cs101-class.org/hub.php
LikeLike
Thanks, I fixed that!
Cathy
LikeLike
In the ‘Shelling Segregation Model’ in the Second Section on Segregation and Peer Effects,
he uses NetLogo to model the segregation effects. The rules (assumptions?) he uses to model the Agents seem overly simple (e.g. – I don’t use those rules when I’m deciding when and where to move). Then he goes on to tweaks the parameters of his variables and sees
how the adjustments correlate to the census data (or other sources).
Is this approach (with overly simple rules) a case of ‘correlation without causation’?
LikeLike
The Shelling Segregation Model in Logo isn’t his. It’s a common demo in academia and replicates Shelling’s initial paper pretty well. At least, that’s what Shelling told me when I first saw it at Harvard.
LikeLike
Coursera looks good. This is on lines of other websites like Udemy, Online courses on WizIQ, Academic Earth and of course Khan Academy. While WizIQ has a LIVE component, Udemy sort of has it, other two and Coursera look to be video. Wonder which way the tide will turn.
LikeLike
Scott is on my dissertation committe, so maybe I can offer an inside opinion. As I read it, this class is intended to be a gateway drug for math and modeling.
It’s targeted mainly at undergraduates who don’t really like math, but suspect that it might someday be useful. There are a lot of people in this category.
The goal is to toss out lots of interesting intuition models and show how a little math can sharpen your intuition. If it works, students will see the value in model building, and future classes in say, optimization, game theory, or econometrics will be a lot more appealing.
@Vic and Franklin: I agree with the idea that bad models can do a lot of damage. IMO, the only solution is to promote math literacy and skepticism so that bad assumptions will get pulled out into the light faster. Teaching simple models seems like the best way to get there.
@Vladimir: I don’t know when the postponement will end, but Scott has mentioned that the whole thing was thrown together at the last minute, plus something about university lawyers slowing it down. (I don’t know which university, but I can see how something like this could be complicated.)
PS – I’m not shilling here. Scott’s on my committee because I like his approach (and have lots of good things to say about him), not the other way around.
LikeLike
Thanks, that’s useful information!
Cathy
LikeLike