Technical Appendix to the paper on violence: What do you do when the data looks like it is power law distributed over a broad range, but cannot be technically a power law? We use dual distribution and transport parameters between one and another.
Violence is from Extremistan, hence requires some more sophisticated tools since LLN works slowly. We see how Pinker’s thesis is bogus. We look at ways to integrate the factual unreliability of historical accounts. We look at transformations to analyze violence using Power law tools since the worst case is bounded at the contemporary population level.
How to look at the risks of Covid vaccines, why they are much lower than you think. We never had a larger monitored sample size in history and it allows events that on average show up later to manifest themselves very early on. Rationale: It takes a long time in a casino for someone to win 8 times in a row. But if 8 billion people played at the same time you would certainly witness a minimum of such events every day.
Explaining path dependence and maximum drawdown. I made a mistake in terminology. What I called max drawdown is a local or “window drawdown” or “peak to the subsequent valley”. The real max drawdown is the one that goes from the peak to 80.
In every age bracket, the vaccinated live longer than the unvaccinated. However as a group, the unvaccinated appear to have a longer life expectancy. This is because the vaccinated tend to be older (hence more likely to die). I explain Simpson’s Paradox in general.
Note: I used the vaccinated/unvaccinated ratio for 50-60 vs 10-20 of Oct 2021, so don’t bug me if it rose since; no effect on the point so long as there is an inequality.