What can be done will be done and so people will build intelligent models for at least the next twenty years. But it's hard for me to see them changing anyone's mind about any major outstanding issue in economics. What comes out will be a function of what goes in. In contrast, regressions and simple models have in many cases changed people's minds....
In my possibly overdogmatic view, economics is most useful when its models are relatively simple and intuitive. We've run out of new models which are simple and intuitive. So the theory game is over. The standard, old data sets have been data mined to death. We're now on to the "can you build/create your own data set?" game. That game can and will last for a long time; in some ways it will favor go-getter extroverts just as the theory game favored introverts.
Take the time to watch this fantastic one-hour video by Michael Welsh (Kansas State) on The Future of Education -- I guarantee it will change the way you think about integrating technology into your teaching.
I fully echo Caron's sentiments on this video. It is superb and highlights the largely untapped potential for using web-based technology (YouTube, Wikipedia, Facebook, Twitter, Blogger, etc.) to take advantage of The Wisdom of the Crowds to turn classrooms into interactive learning laboratories.
At the end of the video, Welsh shows a simulation he and his students perform that mimics the development of an artificial world from 1450 to 2050 and beyond. I would absolutely love to see a similar simulation done in an economics course highlighting how various institutions (legal systems, political systems, cultural norms, etc.), geographies, and resource allocations can lead to vastly different outcomes for people groups over time -- similar to an agent-based model using live people in lieu of computerized agents.
Needless to say, this video has given me a ton of ideas for teaching in the future. I can't wait until I can get a classroom of my own and try to put some of these ideas into practice.
To get a preview of what the video linked to above is like, watch Welsh's earlier video, The Machine is Us/ing Us:
A single ant or bee isn't smart, but their colonies are. The study of swarm intelligence is providing insights that can help humans manage complex systems, from truck routing to military robots.
Simply fascinating!
Some questions this raises -- what other applications are there for these insights? Swarm theory seems to give a powerful explanation for why freedom is so good at generating positive economic results. What else can this describe? Does it help explain the rise of Protestantism though decentralization of religious authority? What application does it have for analyzing the court system of the US? There seems to be a strong parallel between judicial decisions among the various levels of courts in the US and the decision making of bees:
In one test they put out five nest boxes, four that weren't quite big enough and one that was just about perfect. Scout bees soon appeared at all five. When they returned to the swarm, each performed a waggle dance urging other scouts to go have a look. (These dances include a code giving directions to a box's location.) The strength of each dance reflected the scout's enthusiasm for the site. After a while, dozens of scouts were dancing their little feet off, some for one site, some for another, and a small cloud of bees was buzzing around each box.
The decisive moment didn't take place in the main cluster of bees, but out at the boxes, where scouts were building up. As soon as the number of scouts visible near the entrance to a box reached about 15—a threshold confirmed by other experiments—the bees at that box sensed that a quorum had been reached, and they returned to the swarm with the news.
"It was a race," Seeley says. "Which site was going to build up 15 bees first?"
Scouts from the chosen box then spread through the swarm, signaling that it was time to move. Once all the bees had warmed up, they lifted off for their new home, which, to no one's surprise, turned out to be the best of the five boxes.
The bees' rules for decision-making—seek a diversity of options, encourage a free competition among ideas, and use an effective mechanism to narrow choices—so impressed Seeley that he now uses them at Cornell as chairman of his department.
Just as the decisions of bees coordinate around efficient outcomes, so too do the numerous decisions made by various court systems around the country tend towards economically efficient outcomes. This evolutionary, emergent nature is one of the beauties of a common law system and is very analogous to swarm behavior in nature.
Read the whole article! There are many ideas that could emerge from this.
I interviewed Steve Omohundro prior to the Singularity Summit 2007 about his notions of developing artificial general intelligences (AGIs), but he went into far more detail about his views. He applies what he called “microeconomic rationality,” from the work of John Von Neumann and others, to building intelligent, well meaning machines.
Read much more after the link.
See my previous posts on agent-based models (something I hope to make a significant part of my future research):
Go to the ant, O sluggard; consider her ways, and be wise. Without having any chief, officer, or ruler, she prepares her bread in summer and gathers her food in harvest." - Proverbs 6:6-8 (ESV)
What can ants teach us about economics and how to do scientific research? Quite a bit actually!
First, GMU Prof Russ Roberts interviews Stanford Professor Deborah Gordon in this podcast, discussing the similarity between the behavior of ants and that of humans in the economy. They give particular attention to economic concepts such as the division of labor and the emergent order:
Deborah M. Gordon, Professor of Biological Sciences at Stanford University, is an authority on ants and order that emerges without control or centralized authority. The conversation begins with what might be called the economics of ant colonies, how they manage to be organized without an organizer, the division of labor and the role of tradeoffs. The discussion then turns to the implications for human societies and the similarities and differences between human and natural orders.
In Forbes, Nassim Taleb, author of The Black Swan, made some comments I like:
Things, it turns out, are all too often discovered by accident. . . . Academics are starting to realize that a considerable component of medical discovery comes from the fringes, where people find what they are not exactly looking for. It is not just that hypertension drugs led to Viagra or that angiogenesis drugs led to the treatment of macular degeneration, but that even discoveries we claim come from research are themselves highly accidental. They are the result of undirected tinkering narrated after the fact, when it is dressed up as controlled research. The high rate of failure in scientific research should be sufficient to convince us of the lack of effectiveness in its design. If the success rate of directed research is very low, though, it is true that the more we search, the more likely we are to find things “by accident,” outside the original plan.
If the success rate per test is low, a good research strategy is to start with low-cost tests. Ants do this: They search with low-cost tests (single ants), exploit with high-cost tests (many ants). I don’t think the need to use different tools at different stages in the scientific process is well understood. John Tukey used the terms exploratory data analysis and confirmatory data analysis to make this point about data analysis but distinguishing exploratory and confirmatory experimental design is much less common.
Read my previous posts on The Black Swanhere and here and think twice the next time you see an ant moving around. Turns out we may have quite a bit to learn from them.
You can run an agent-based model of ants foraging for food here. This is one of the models that got me interested in considering agent-based models as a tool for economic research.
Axtell is one of the leaders in this field of research. In 1996, Axtell cowrote a seminal work on artificial societies titled “Growing Artificial Societies: Social Science from the Bottom Up,” with Joshua Epstein of the Brookings Institution. In the book, Axtell and Epstein present a computer model with which they begin to develop a bottom-up social science in a land known as Sugarscape.
As various changes in environment or agents are introduced, data on differing outcomes are produced. What the authors found is that “fundamental collective behaviors such as group formation, cultural transmission, combat and trade are seen to emerge from the interaction of individual agents following a few simple rules.”
Currently, there are three main areas where multiagent systems modeling affects policy: traffic; propagation of disease, particularly since Sept. 11, 2001; and military tactics. Axtell says that teaching MBA students to build models of entire companies with these agent-based techniques is definitely on the horizon. Using this type of modeling to address institutional dilemmas could one day help corporations analyze problems and determine more effective solutions than might have been possible in the past.
“Our models essentially enhance the equation-based methodology (EBM), which is, in a manner of speaking, the forerunner of the modeling we do,” Axtell says. “But EBM misses all the action around the average, and, in some cases, that action matters a lot. Predicting extreme events is what’s most important.”
I took Dr. Axtell's Agent-Based Modeling class this past semester and must say it was one of the most interesting and thought-provoking classes I have taken thus far in my PhD program. I enjoyed it so much I am now taking an intro to computer science course this summer to learn better programming skills to use agent-based models in my research.
For any of my fellow PhD students at GMU, I cannot recommend Dr. Axtell's class highly enough. I also recommend Eric Beinhocker's excellent book, The Origin of Wealth, as a good layman's introduction to many of the concepts we covered in the class. (Download chapter 1 here. [PDF])
Here is a PowerPoint presentation I gave on The Origin of Wealth giving a brief overview of the breadth of topics covered in the book:
The agent-based modeling approach strikes me as a much fuller and richer way to model human behavior than traditional mathematical analysis. When I was working as an engineer, I got invovled with doing Monte Carlo simulations to analyze material failure in steam turbines. Agents are an even more powerful tool for social science research that allow the modeling of heterogeneity of individuals, randomness of events, congnitive limitations, social influence of neighbors, etc. in a way that conventional analysis cannot address. Professor Larry Iannaccone and Michael Makowsky have even used this approach to model the effects of religious commitment.
If you can't tell, I'm very excited by the development of these tools and hope to make them a big part of my analytical toolkit.
Tonight, I've been running a bunch of simulations for a project for my agent-based modeling class. While the models were running, I listened to this intriguing podcast with Russ Roberts interviewing NassimTaleb, author of The Black Swan.
Taleb discussed the differences between events that follow Gaussian distributions (normal distributions) and those that follow power law distributions. Most of conventional statistics is based on normal distributions while many events in life actually follow power law distributions, making many of the tools used for data analysis and risk-management very flawed. This affect fields as diverse as medical research (according to Taleb, 80% of epidemiological studies are not replicable), finance, and economics.
To get at the distinction between a normal distribution and a power law distribution, think about comparing the height or weight of people to their wealth. Height and weight of adult males can certainly range from short to tall, but you'll never meet a 6-inch-tall man or a 50-foot tall giant. The constraints on this range make using a normal distribution an appropriate representation for the variation in this population. On the other hand, you might encounter someone from Botswana who makes $300 per year while Bill Gates is worth billions. The distribution of wealth follows a power law, with a high number of low-wealth people and a small number of high-wealth people.
A problem arises when researchers and analysts inappropriately use statistical tools designed for normal distributions for things that follow a power law. Stock markets, earthquakes, book sales, etc. all are examples of things that are sometimes prone to unforeseeable rare events (both positive and negative). The troubling thing about power laws is that they are unpredictable. One of the key insights of Taleb is that it is important to know what we don't know and once we do, it is possible to plan for the unexpected. We often get into our biggest trouble when we assume we know much more than we do.
Listen to this fascinating podcast and be sure to check out Taleb's book. I ordered a copy on Wednesday and it should get here later this afternoon.
In the November Scientific American, Stuart Kauffman offered a bias theory to explain why economists show little interest in his "complexity" economics:
Unexpected change bedevils the business community endlessly ... Economists have so far not been able to offer much help to firms trying to be more adaptive. Although economists have been slow to realize it, the problem is that their attempts to model economic systems focus on those in market equilibrium or moving toward it. ... The path to maximum prosperity will depend on finding ways to build economic systems in which new niches will generate spontaneously and abundantly. Such an approach to economics is indeed radical. It is based on the emergent behavior of systems rather than on the reductive study of them. It defies conventional mathematical treatments because it is not prestatable and is nonalgorithmic. Not surprisingly, most economists have so far resisted these ideas. Yet there can be little doubt that learning to apply these lessons from biology to technology will usher in a remarkable era of innovation and growth.
The claim seems to be one of academic tool overconfidence: academic communities are biased toward the conceptual tools they use the most, and against approaches that rely on other tools. Now there are large coordination payoffs from academics using similar tools, since this lets them compare and build on each others' results. And this must put pressure on any given researcher to use tools similar to those used by others in his field. But it is far from obvious to me that this constitutes a bias.
I don't think this constitutes a bias so much as the effects of investments of human capital in specific tools (convential economic methodology). I think economics is like any other field and that new techniques and tools take time to get build, adapt, and get accepted. I do think there may be structual issues within academia that make it resistant to rapid change (not all aspects of this are a bad thing -- imagine if every new, untested theory was instantly accepted by everyone). I thin it will take time to reorient the profession towards more adaptive models, assuming this line of research bears fruit. (There is also probably a public choice story to be told here as well.)
I think the social sciences (including economics) may already trending in the direction of adaptive models. I see these tools as a compliment rather than a substitute for current economic methods. They may allow for a much better testbed for conventional (and unconventional) economic thinking.
Coincidentally, I am taking a course in Agent Based Modeling (ABM) this semester with Robert Axtell (who recently came to GMU). It looks to be an exciting class and I am impressed with the applications of ABM to many fields of study. I'm sure I'll be blogging more about this in the near-future. In the meantime, you can read more about ABM here.
Tyler Cowen on agent-based modeling:
