Typically when we explore the outputs of agent-based models we do a parameter sweep: change the values of the parameters according to some pre-defined scheme and take measurements of the outcome from each of those parameter sets. Typically our models are stochastic so we do multiple trials of each set of parameters. Similar approaches are used for many other modeling techniques (including deterministic ones) to measure sensitivities and generate comparative statics. This post focuses on an alternative to the standard parameter sweep that intelligently searches through the parameter space to find the regions where the outcome is most variable, i.e. the interesting parts.

» Read More

Hello blog readers. I've been forced to keep commenting disabled on this blog for the past few years because the spam bots are too vigilant and hit the posts with hundreds of viagra adds every night if I don't. I just ran a two-day test to see if I could reopen commenting, and after deleting over 400 spam messages I concluded that I cannot. But I really wish I could because getting feedback is a large motivating reason for writing these posts in the first place. If any of you out there have any ideas on how I could set things up to have comments enabled without getting plastered with spam messages then I'd greatly appreciate it if you'd drop me a line at bramson@umich.edu and let me know your idea. Thanks in advance.

The newest version of Netlogo includes and much asked for (at least by me) feature: transparency! It applies to turtles (and it also applies to links and pens) and only for 2D models, but it's a great addition to the program. That said, there are some eccentricities in how one must use Netlogo's transparency features and I've worked out some these so you don't have to. What follows is some procedures you can include in your code to use transparency in the way that you should be able to use transparency...just like another turtle variable such as size or color. In addition to the examples and explanations below you can download my personal transparency code example and use that to freely steal my code for your own use.

» Read More

There is a lot of talk in social science departments about the correlation/causation gap and to what degree can causal claims be justified from association data. Those discussions might be clichéd by now, but they are still important and furthermore address issues in decision theory, policy, and experiment design that are more relevant now than ever. But this post is not about the distinction between causation and correlation; it's about the distinction between causes and reasons. Both causes and reasons factor into explanations for social phenomena, but their roles ought to be different because causes and reasons are different. The distinction and their differences are sometimes less than fully appreciated and I've got a few things to say about that here and now.

» Read More

The topologies of 2D agent-based models are almost completely dominated by bounded planes and torus surfaces (aka: wrapping edges, aka: periodic boundaries). These may be reasonable approximations for some systems, but certainly not for all. There are at least few things I’d like to model that occur on the surface of something roughly spherical (e.g. on the Earth). Spheres are topologically like bounded planes in that neither has holes and both can be laid flat. But they are also like tori in that they have periodic boundaries as surfaces of three-dimensional objects. In order to facilitate a growth in popularity of sphere (or ellipsoid) models, I present a technique (and eventually code) for generating such models in Java or NetLogo.

» Read More

There are reasons to doubt that the so-called naturalistic fallacy is really a fallacy, but at this point I want to accept the major points of Hume's version of the open question argument and see how that actually helps us in comparing metaethical theories. The idea is that even if everybody agrees on the facts regarding our evolutionary history we might still disagree about which features of that story convey or produce moral value. We can use this difference to distinguish moral theories that otherwise tell the same story. It also lays bare those aspects of a evolutionary history that might do the metaethical job and thus reveal whether any such story will do the job.

» Read More

A famous story in decision theory is the ass between two identical bales of hay that, because he does not have any reason to choose one or the other, dies of starvation. Other stories such as Death in Damascus and the Newcomb Problem are more complicated (see below), but also result in a deliberative state in which no unique outcome is deemed appropriate. And then there are all the mixed strategy equilibria in game theory. These are arrangements where the other player has randomized over her strategies in such a way as to make you indifferent among all of your own. In all these cases and others lacking a unique decision recommendation there isn't a thing to do, you can simply pick one or select randomly or whatever...no outcome is better than the others. Here I want to briefly consider how various indifference cases split when we consider the possibility that one could create clones and pursue multiple actions simultaneously. Depending on whether such clones are indifferent to changing places with other clones (and other such considerations) we can categorize the states as different kinds of indifference.

» Read More

There have been many attempts to define culture, each having its own spin reflecting what work the definition needs to do. My favorites are Boyd & Richerson's and Page & Bednar's, though they are quite different from each other. What the existing definitions have in common is that they are attempts to capture what culture is rather than what culture does. My claim is that culture is properly understood as systematic patterns in how things are done; and my focus is on measuring cultural distances rather than simply identifying cultural components or providing culture-based explanations. My measurement scheme analyzes similarities in the differences of how things are done: the "same difference" criterion. This measurement does not help provide an explanation for culture and their differences, but it does identify which behavioral features need a cultural explanation. A sketch of the technique follows.

» Read More

There are several approaches to the genetic algorithm (GA) technique (e.g. variations in parameters, birth-death timing, encodings) and these differences typically affect the performance and outcome of the technique. There are some cases where the appropriateness of one variation is clearly better than others, but many more cases of a technique being chosen merely by convention or because something still must be done when there are no differentiating reasons. Part of the problem is that the differences generated by the variations in GA approach are things which are very difficult to measure with existing metrics: dynamical properties relating to the number and location of attractors, their relative basin sizes, the speed of convergence, etc. These are precisely the sorts of properties that I developed the tipping point methodology to be able to measure and so bringing it to bear on a few GA variations of a common problem may yield some valuable insights.

» Read More

There is a common inclination in cases of deep uncertainty to apply probabilities uniformly across all possible future states, despite this inclination being unsupported by any theoretical principle. The notion may be derived from (i.e. be applied analogously from) the behavior rules guiding distributive justice wherein it is accepted that, if differentiating reasons are not to be found, then dividing a resource (or requirement) equally among possible recipients is prescribed. The rule internalized for the historically common cases of distributive justice in past human society may be providing the uniform probability distribution intuition in the decision calculation procedure. This case might therefore be an instance of the evolutionary history of humans generating a reliable bias in modern human decision problems and thus merits consideration for institution design.

» Read More