Computer Generated Farmers: Information and Nash Equalibrium

Introduction

Economic actors in general are greedy. They will try to better their outcome using information they acquire, mimicing the strategies of anyone who is doing better than them. Our question was, given a simple economic system and actors following a simple rule: do as other who are doing better than you are doing, will this system ever reach a reasonably stable nash equalibrium? Will it ever become pareto optimal? And will the economic actors be happy? In order to answer these questions, we built a computer generated simulation program and monitored how certain factors influenced the system.

Objective

We used computer gerated farmers to simulate a special economic system, in which each farmer had a limited number of friends who they could get information from to try to better their strategic outcome.

Setup

We will generate 100 digital farmers, each having a production capacity of 1, choice of 2 crops to make (corn and rice), and a certain number of friends who they can retrieve production information from. We will call the amount of corn a farmer makes alpha. Since each farmer has a production capacity of 1, naturally the amount of rice this farmer makes will be 1-alpha. From each friend a farmer has, the farmer can retrieve two information. One is the total profit the friend made in the previous production period, and the other is the friend's alpha. The profit the farmer makes will be determined by a rational function expressing the relationship between profit per crop made versus total amount of crop made, which in general form would be P=a/x. Therefore, each profit function would look like this:

File:Function.png You have to be logged into the wiki to upload a image.Gribble 15:34, 12 Apr 2006 (EDT)