gbepfit1
Fit a symmetric power exponential density
Usage
gbepfit [options]
Overview
Fit symmetric power exponential density using maximum-likelihood or the method of moments.
Options
-
-O -
output type (default 0)
0
|
parameter b a m and log-likelihood |
|---|---|
1
|
the estimated distribution function computed on the provided points |
2
|
the estimated density function computed on the provided points |
3
|
parameters b a m and their standard errors |
-
-x -
set initial conditions b,a,m (default 2,1,0)
-
-m -
the mode is not estimated but is set to the value provided
-
-s -
number of intervals to explore at each iteration (default 10)
-
-V -
verbosity level (default 0)
0
|
just the final result |
|---|---|
1
|
headings and summary table |
2
|
intermediate steps results |
3
|
intermediate steps internals |
4+
|
details of optim. routine |
-
-M -
active estimation steps. The value is the sum of (default 7)
1
|
initial estimation based on method of moments |
|---|---|
2
|
global optimization not considering lack of smoothness in m |
4
|
local optimization taking non-smoothness in m into consideration |
-
-G -
set global optimization options. Fields are step,tol,iter,eps,msize,algo. Empty field implies default (default .1,1e-2,100,1e-3,1e-5,3)
-
-I -
set local optimization options. Fields are step,tol,iter,eps,msize,algo. Empty field implies default (default .01,1e-4,200,1e-4,1e-5,5)
-
-F -
input fields separators (default " \t")
-
-h -
this help
The optimization parameters are step initial step size of the searching algorithm
- tol
-
line search tolerance iter: maximum number of iterations
- eps
-
gradient tolerance : stopping criteria ||gradient||<eps
- msize
-
simplex max size : stopping criteria ||max edge||<msize
- algo
-
optimization methods: 0 Fletcher-Reeves, 1 Polak-Ribiere, 2 Broyden-Fletcher-Goldfarb-Shanno, 3 Steepest descent, 4 Nelder-Mead simplex, 5 Broyden-Fletcher-Goldfarb-Shanno v.2, 6 Nelder-Mead simplex ver. 2, 7 Nelder-Mead simplex rnd init.
Examples
- gbepfit -m 1 -M 6 <file
-
estimate a and b with m=1 and skipping initial method of moments estimation
Releases
Unchanged in 3 releases since Debian 11 (Bullseye).