Sunday 17 August 2014

Mathematicians formalize a method to measure how wrong we are

Science Focus

original post »

Statistical analysis is an indispensable tool in most of today’s science. John Turkey, one of the most important statisticians of the second part of the 20th century and a person who coined the term “software” once said that “’the best thing about being a statistician is that you get to play in everyone’s backyard.” Despite being used in almost every field of research ranging from quantum mechanics to sociology and meteorology, conceptually it can easily regress into a kind of argumentative loop, for given a statistical account of the way things are, one may ask what is the probability that this particular account is true? Although it is a question of great philosophical depth, pondered by thinkers working in the field of epistemology, it is a priori impossible to provide a probability that any model is true, because the model itself is an attempt at truth or its possibly successful approximation. But, doing what is available, scientists have devised methods to measure the possible distance between true (statistical) state of the world and the description that is assigned to it via some measurement process. One of such methods is called Kullback-Leibler divergence and it measures the amount of information gained

The post Mathematicians formalize a method to measure how wrong we are has been published on Technology Org.

 
#physics 
 » see original post http://feedproxy.google.com/~r/TechnologyOrgPhysicsNews/~3/IAF3dgugT4s/
See Zazzle gifts tagged with 'science'

No comments:

Post a Comment