Copyright | (c) 2015 Mihai Maruseac |
---|---|
License | BSD3 |
Maintainer | mihai.maruseac@maruseac.com |
Stability | experimental |
Portability | portable |
Safe Haskell | None |
Language | Haskell98 |
The Laplace distribution. This is the continuous probability defined as the difference of two iid exponential random variables or a Brownian motion evaluated as exponentially distributed times. It is used in differential privacy (Laplace Method), speech recognition and least absolute deviations method (Laplace's first law of errors, giving a robust regression method)
Documentation
data LaplaceDistribution Source #
Constructors
:: Double | Location |
-> Double | Scale |
-> LaplaceDistribution |
Create an Laplace distribution.
laplaceFromSample :: Sample -> LaplaceDistribution Source #
Create Laplace distribution from sample. No tests are made to check whether it truly is Laplace. Location of distribution estimated as median of sample.
Accessors
ldLocation :: LaplaceDistribution -> Double Source #
Location.
ldScale :: LaplaceDistribution -> Double Source #
Scale.