Feb 13, 2014

The P-Value Was Never Meant To Be Used The Way It's Used Today

I have just came across Scientific method: Statistical errors  published Nature's website. I found it a very good article about the misuse and misinterpretation of p-values and about scientific evidence in general. It says, among other things, that “the P value was never meant to be used the way it's used today.”. I was not aware of the history of introduction of this measure and it was interesting thing to read. I was not surprised by some of the points about the quality of scientific research as it has been covered a lot by Taleb, Goldacre and Ioanidis to mention a few but it was nice to see the connection between these two. The author suggest an approach to curb the p-hacking effects in science but considering the nature of science studies these day I do not believe it is applicable for the majority of research. There is other good thing to learn in the article, so please read it yourself.

Feb 12, 2014

Ban GDP as A Driver in Regressions!

If there is one thing in analytics that grinds my gears it is the overuse of GDP as a driver in regression. If it were up me, I would ask the government and academy of science to issue a book with all approved regression drivers and distribute it as a standard repository for all knowledge and wisdom there is related to regression. Then one copy of this book and one copy only would include GDP. This copy will be in Klingon and as a precautionary measure will be booby trapped with a potent explosive and buried 100m in the ground. On the Moon. On the dark side of it.