Intel® Fortran Compiler
Build applications that can scale for the future with optimized code designed for Intel® Xeon® and compatible processors.

GPD

JohnNichols
Valued Contributor II
202 Views

The fatality model has a component that is GPD - see equations.  I need to be able in Fortran to determine a million random numbers that fit the GPD data for a given theta and lambda - has anyone done this sort of thing before. - can I use a transformation onto a standard distribution 

Capture.JPG

0 Kudos
3 Replies
Andreas_Z_
New Contributor I
202 Views

The 'prob' library may be of use: https://people.sc.fsu.edu/~jburkardt/f_src/prob/prob.html; it offers several probability density functions and associated CDFs, not sure about GPD.

andrew_4619
Honored Contributor II
202 Views

IMSL has some Generalised Pareto functions.

 

JohnNichols
Valued Contributor II
202 Views

The GPD is a challenging Probability Distribution -- I was thinking about how do the maths people do the transformation from a uniform PD to a Gaussian -- the procedure must be codable.  Also I was wondering if we could treat the problem like a refraction thru a lens -- the output form the lens i the GPD numbers and the input is uniform random numbers.

There was 2400 deaths yesterday - the virus is a bastard.  

Reply