In my "stress model" I assumed a correlation between violent crime rate and the stress in a population. Here, "stress" is defined as the difference between the kind of life a person has and the one the person wants to have (the "comfort zone"). This difference can be theoretically modeled as a distance within a fixed range of possibilities. Assuming there is a level of stress beyond which someone turns violent, the crime rate is proportional to the number of people in a population who are beyond that limit (specifically, the number per 100,000 population, divided by the average lifetime of people in the population).

Distance traveled (and therefore stress) can be modeled by randomly choosing values of a variable (called "adjusted power") that measures the direction and fraction of the distance between a person and their goal that they can travel in an arbitrary unit of time (such as a lifetime).

Two models were created based on different assumption about how the violent crime rate relates to stress. Model 1 assumes that the part of the population beyond a certain stress level will commit violent crime. Model 2 assumes that the violent crime rate is proportional to the average stress in the population.

These models have been superceded by the "Power" model, based on a comparison of "happiness" with several variables. "Adjusted power" appears to be related directly to happiness, and violence actually INCREASES with both. See Happiness and Power for details.

---

For more discussion, see Comfort and the related blog entries starting with "Security."