|
|
|
|
|
|
by jhbrown
1247 days ago
|
|
|
You're exactly right! Under the hood, DoomCheck actually uses a log-logistic distribution -- https://en.wikipedia.org/wiki/Log-logistic_distribution -- which it fits using the range endpoints as 10% and 75% respectively. Then it does a Markov simulation to combine them across the whole project. |
|
|