stable: Fitting Stable Distributions

In christopherggreen/cggmisc: Christopher G Green's Miscellaneous Functions

Description

Fit alpha-stable distributions using maximum likelihood estimation.

Usage

fit.mle.stable(obsvals, method = c("nlminb","optim"),
  mean.exists=TRUE, robust.location=FALSE, strace=FALSE)
likelihood.stable(x, y, obs, mean.exists=TRUE) 

Arguments

obsvals	A vector of observations.
method	Which optimization routine should be used to compute the MLE? Current options are nlminb and optim. Default is nlminb.
mean.exists	Should it be assumed that the underlying stable distribution has a finite first moment (i.e., that alpha > 1)? By default it is assumed that alpha > 1, and this condition is enforced in the fitting process.
robust.location	Should a robust location estimate be used to compute the initial estimate of the MLE? Default is FALSE.
strace

x
y
obs

Details

fit.mle.stable fits an α-stable model using maximum likelihood estimation. The distributional model in use here assumes that the random variable X follows a distribution of the form

X \sim

alpha is the tail index parameter of the stable distribution. fit.mle.stable uses the likelihood function provided by likelihood.stable.

Value

alpha
beta
sigma
mu
hessian

Author(s)

Christopher G. Green [e-mail]

References

Green, C. G. (2005) Heavy-Tailed Distributions in Finance: An Empirical Study. Qualifying Exam, Department of Statitics, University of Washington. Available from http://students.washington.edu/cggreen/uwstat/papers/computing_prelim_2005_green.pdf

See Also

fit.mle.t

christopherggreen/cggmisc documentation built on May 13, 2019, 7:04 p.m.