I'm doing a wee bit of analysis looking at app downloads.
They obviously follow a power-law distribution based on search rank, and it's trivial to create an equation that matches that eg: D=pR^-1.2
I'm having trouble picturing the behavior in my head that would lead to this kind of distribution though:
A variable will fall into a power law distribution if it is an exponential function of a second variable that is distributed exponentially.
I can show that number of eyeballs looking at an app are probably exponentially declining, assuming a constant probability of choosing an app and people moving down the list. (so if it was 50% chance to choose the app you were looking at, half the guys would grab the first app and stop looking, the other half would look at then 2nd app, 50% of that half would grab that and the remainder move to the next app and so on)
If the probability of choosing the app is also exponential, then you get the right power law behavior. Is it fair to assume quality/relevance drop off exponentially in a search result ranking?
[Maths] - Power law creation
Moderator: Alyrium Denryle
[Maths] - Power law creation
"Aid, trade, green technology and peace." - Hans Rosling.
"Welcome to SDN, where we can't see the forest because walking into trees repeatedly feels good, bro." - Mr Coffee
"Welcome to SDN, where we can't see the forest because walking into trees repeatedly feels good, bro." - Mr Coffee