History lessons in data science

 

By D. Thomakos

Sage advice on the application of statistical methods from a 1937 book by Harold Davis and W. F. C. Nelson. Davis was a pioneer in the introduction, research and application of mathematical statistics and econometrics in economic time series and economic data. Click on the Cowles Foundation webpage and get his works, there is a lot to be learned from history - if nothing else, read below and see how back in 1937 it was already understood that the assumption of a normal distribution was not applicable everywhere. Many years before some modern proponents of heavy-tailed distributions came to the foreground and denounced normality, the excerpt below was placed as a forewarning to the data scientists of the day. Read your history books before getting your hands on the math or the code, for history will improve your good judgement!

Davis and Nelson