A good article on the importance of looking at Paretian distributions for visualizing data. The main point that the author makes is that Paretian distributions are important tools for visualizing important upcoming trends in the data. Guassian distributions have a tendency to show important future trends as unimportant outliers in the data set, that is future trends will only appear outside of the Bell curve until it is too late to take any action. However, using a properly created Paretian visualization, the importance of data points that fall outside the norm, but which are attracting lots of attention or have many 'links' will be visually apparent.

Bookmark/Search this post with:

delicious | digg | reddit | furl | technorati