Wednesday, January 11, 2006

Likelihood Statistics

I was just reading this great post from Mike the Mad Biologist (I know ... a much more original name than my nom de plume).

In this entry, he discusses Falsification (as in the theory of scientific theories espoused by Karl Popper) and likelihood Statistics (a much more realistic description of how science works).

From his post:
The difference between the two approaches is that falsification tests data against an a priori model, while likelihood uses the data to build the most likely model given the existing data. The strength of likelihood is that it does not assume how the world works. It also allows you to judge the relative likelihood of different models (or processes). The disadvantage is a garbage-in garbage-out problem: 92% of the time, an unbiased coin would yield an observed ratio that is not 50:50.
It's worth reading the whole entry - and if the detailed discussion frightens you, Mike also incorporated cuddly pictures of baby pandas and puppies.