Sure, why not.
Take a complete selection of prices,without changing frequency, take them from a long sequence of prices. Then group them into five sets, each set having nearly the same mean price to variance price among them, more or less. The sets of prices should be grouped by mean.
Then pick a base, 2,3 or e; and a noise exponent near one, and set them such that the sum 1/B^(noise*k), k integer from 1 to 5, is almost the mean -rice for each group. The rate of noise change relative to s*k gives you the temperature or transaction rate, as a relative value. The production graph computes one unit of GDP for each set of k*n steps. Just feed the bottom of the production graph with prices as they come. When 2 or 3 prices meet at a node, step that combined price forward, like a Petri net (My engineering professor) I will have to look him up, he was big into Petri nets.
No comments:
Post a Comment