Yeah.. I get that it is a hard problem, and that oversimplified models are not likely to yield particularly useful information. I’m not familiar enough with economics or financial markets to properly judge what would or would not be practical to predict with computer simulations. I’m just saying that it should be possible to gleam some useful insights from examining the behavior of numerical models before one attempts to unleash yet another money making hording derivative financial product on the market.

On a related note, I’ve just finished reading through the latest issue of SIAM News (Jan/Feb 2009 – not online yet). There is an article describing how there has been a significant amount of dispersions cast at the so-called ‘quants’ who came up with financial forecasting models that were used rather carelessly by day traders. Apparently, there has been no shortage of highly qualified PhD types working on financial modeling and risk analysis. Sadly, it seems as though there was actually a tremendous shortage of managers who would actually listen to the analysts and pay attention to what the models were trying to telling them.

Discussions of how the banks got themselves into the positions they did often center around the models, and the typical question is, “Were the models wrong?” or “Were the inputs to the models wrong?” Whatever answer one chooses to give, the larger failing is human, in how the risk models were developed, understood, managed, applied, and, for the most part, ignored.

The full article should appear online here sometime in the next couple of weeks.

This applied equally well to the start of the space race. In 1958, Lyndon Johnson, then majority leader of the Senate, warned that if the Soviets controlled space then they “would have the power to control the earth’s weather, to cause drought and flood, to change the tides and raise the levels of the sea, to divert the gulf stream and change temperate climates to frigid.” No sensationalism there! Read that quote again, just for effect. Anything line used to sell the Iraq war or the stimulus package is quite mild in comparison.

Incidentally, just to clarify, the invasion of Iraq was never about fighting Al Qaeda, and Bush & Co. never claimed that it was. The purpose was to reshape the geopolitical paradigm in that region to the West’s favor. That isn’t automatically a bad idea, but it can (and did) suffer in its execution. Still, the end result is that Iraq has a better government now than any in that region except Turkey and Israel. Was it worth the effort? That’s a different debate, but it does resemble what the future may bring with the stimulus bill: The stimulus may “work,” but almost certainly not very well, and definitely not as well as saner alternatives.

“Given a sufficiently accurate mathematical models, computational simulations are amazingly powerful. But they can also be used to formulate and test new models. Physicists do this stuff all the time. Why can’t economists?”

I think this relates to the state of economic theory in the world today. From what I’ve seen, most economic theories start out with, “Assume perfectly rational actors and perfectly efficient markets…” Professional economics simply isn’t operating at a level of detail that would allow accurate computer simulations. Imagine doing a CFD simulation of orbital re-entry assuming the atmosphere always behaves according to the subsonic ideal gas law.

I have heard of behavioral economists who are trying to get accurate, detailed theories of people’s actual economic behavior through experiments and observation. This is a good trend, but I think it hasn’t infused far enough into mainstream economic circles. And the behaviorists are probably also just starting to learn the basics. It will take time to build up theories of more complex group dynamics.

As for the hardness of the problem itself, there is something to be said for that because the problem is chaotic. It’s like trying to predict the weather where a small change in initial conditions can result in a large change in outcome.

But we do have weather simulations that can predict at least a while into the future, and they are getting better all the time. I agree it’s mostly a cop-out to claim that economic simulation can’t be done. It protects those with beautiful theories that aren’t useful in the real world.

John Hare wrote:

“One of my many buttons is the people that don’t know the difference between creating wealth and obtaining it, and don’t care”

I agree with this, and I think one of the causes of this problem is the current oversimplified economic theories that say it is impossible to obtain wealth without creating it, therefore if you got some money you must have deserved it.

None taken, but irrational does not equal random. In fact I’m speaking of mass irrational behavior such as Joe Sixpacks becoming a real estate investors at the peak of the real estate bubble, or people buying tulips 2 centuries ago. Market cycle psychologies go through start at fear to invest, cautious optimism, mania, and then back to fear. Those psychological variables of mass behavior can be approximated based on past cycles. This is why a small number of people saw “it” coming in 1929, and more recently over the last few years.

As for Obama’s first actions, I’m not surprised that he likes real torture enough to close the Club Gitmo and keep the renditions. He is, after all, a Democrat. Now the terrorists are not going to have the option of gaining a few pounds in a resort in Cuba. Maybe it’s all to the better.

No offense, but I think that is a cop-out. We are not talking about a completely random or chaotic process. There are elements of that in any system that contains large numbers of interacting variables. However, by and large, individuals do not act randomly. More than likely they will have a very high probability of taking certain actions under certain conditions (see game theory).

My concern at the moment is not so much with the actions of individuals, but with the behavior of the system as a whole and its overall robustness. How do we ensure that there are sufficient buffers built into the system to absorb transient shocks? More importantly, how do we identify so-called ‘toxic assets’ sooner so that they may be appropriately dealt with before they precipitate a crisis?

Exactly what freedoms are we being scared into giving up? Under Bush this question was easy to answer: the right to a trial by your peers, the right to due process, the right to be free from illegal search and seizure, etc. were all abridged under the premise that we would not be safe without abridging them. What freedoms are we being scared into giving up by paying to rebuild bridges?

Do you have an answer, or are you just “fear-mongering”?