Referring to the one who predicted the 2008 crash, I thought his cycle theory matched Shannon channel theory, Yes.
But first, its biology meets Shannon, biology gives us 18 years to grow a family, a cycle fixed by biology.
Shannon tells us our only other free variable is th neighborhood size. How many families per neighborhood give is the iLog(i) we need to match local economic conditions (maximize mutual sector entropy). Think in terms of economies of scale, children of a generation band together, to reduce costs. What is the optimum group size.
Computationally, if we could look at cities and perform Huffman encoding on all deliveries over time and place, we could find the best median for neighborhood sizes.
So the 18 year cycle, in Shannon terms, and the imprecision allocated, yield large neighborhood sizes. Channel events look volatile because the best economies of scale are large neighborhoods.
No comments:
Post a Comment