Computation of the incomplete gamma function ratios and their inverse

  title={Computation of the incomplete gamma function ratios and their inverse},
  author={A. Didonato and A. H. Morris},
  journal={ACM Trans. Math. Softw.},
An algorithm is given for computing the incomplete gamma function ratios <italic>P</italic>(<italic>a</italic>, <italic>x</italic>) and <italic>Q></italic>(<italic>a</italic>, <italic>x</italic>) for <italic>a</italic> ⪈ 0, <italic>x</italic> ⪈ 0, <italic>a</italic> + <italic>x</italic> ≠ 0. Temme's uniform asymptotic expansions are used. The algorithm is robust; results accurate to 14 significant digits can be obtained. An' extensive set of coefficients for the Temme expansions is included. An… Expand
Algorithm 708: Significant digit computation of the incomplete beta function ratios
A transportable Fortran subroutine based on this algorithm is currently in use and is accurate to 14 significant digits when precision is not restricted by inherent error. Expand
Efficient and Accurate Algorithms for the Computation and Inversion of the Incomplete Gamma Function Ratios
Algorithms for the numerical evaluation of the incomplete gamma function ratios and inversion methods for solving the equations with positive values of a and x are described. Expand
The Inverse Nakagami-m Distribution: A Novel Approach in Reliability
The inverse Nakagami-m (NK) distribution is proposed, which is the reciprocal of the NK model that plays an important role in the general area of the communications engineering and reliability system. Expand
Computing the Incomplete Gamma Function to Arbitrary Precision
It is shown that the incomplete Gamma function Γ (a, z) can be computed to P digits in at most O(P) long multiplications uniformly in z for Re z > 0. Expand
Uniform asymptotics for the incomplete gamma functions starting from negative values of the parameters
We consider the asymptotic behavior of the incomplete gamma functions $gamma (-a,-z)$ and $Gamma (-a,-z)$ as $atoinfty$. Uniform expansions are needed to describe the transition area $z sim a$, inExpand
Efficient and Accurate Parallel Inversion of the Gamma Distribution
  • T. Luu
  • Mathematics, Computer Science
  • SIAM J. Sci. Comput.
  • 2015
A method for parallel inversion of the gamma distribution is described, which has accuracy close to a choice of single- or double-precision machine epsilon for random number generation in Monte Carlo simulations where gamma variates are required. Expand
Computation of auxiliary functions in STO molecular integrals up to arbitrary accuracy. I. evaluation of incomplete gamma function En(x) by forward recursion
A new inequality and an approximate relation of incomplete Gamma function En(x) are derived in this paper. Applying both to simplify the error analysis on the recursive algorithm of En(x), weExpand
Efficient time/space algorithm to compute rectangular probabilities of multinomial, multivariate hypergeometric and multivariate Pólya distributions
  • R. Lebrun
  • Mathematics, Computer Science
  • Stat. Comput.
  • 2013
This work proposes a new algorithm that allows to perform the computation of arbitrary rectangular probabilities in the most general case and its accuracy matches or even outperforms the accuracy exact algorithms when the rounding errors are taken into account. Expand
Algorithm 969
The algorithm combines the use of series expansions, Poincaré-type expansions, uniform asymptotic expansions, and recurrence relations, depending on the parameter region, for computing the incomplete gamma function γ*(a, z). Expand
Approximations for the inverse cumulative distribution function of the gamma distribution used in wireless communication
The Quantile mechanics approach was adopted to transform the probability density function of the gamma distribution to second-order nonlinear ordinary differential equations (ODEs) whose solution leads to quantile approximation, which yielded values closed to the R software values via Monte Carlo simulation. Expand


Computation of the Incomplete Gamma Function Ratios
A method is given for computing the incomplete gamma function ratio, P and its complement Q(a, x) for all real arguments a 0, x or = 0, which is efficient, yields both P and Q correctly to within 1 unit in the twelfth significant digit. Expand
Algorithm 542: Incomplete Gamma Functions [S14]
An algorithm implements the procedure developed in [1], and the complete listing is available from' the ACM Algorithms Distribution Service (see inside back cover for order for~). Expand
An asymptotic expansion for the upper percentage points of the ²-distribution
An asymptotic development is given for estimating the value of the variable X for which the x2-distribution 2 V = 1 r10tv/2l1etdt Q(X ) F(v/2) rX2/2 assumes a preassigned value oc, in the regionExpand
A Computational Procedure for Incomplete Gamma Functions
We develop a computational procedure, based on Taylor's series and continued fractions, for evaluating Tncomi's incomplete gamma functmn 7*(a, x) = (x-"/F(a))S~ e-~t'-ldt and the complementaryExpand
Rational Chebyshev approximations for the error function
This note presents nearly-best rational approximations for the functions erf (x) and erfc (x), with maximal relative errors ranging down to between 6 X 10-19 and 3 X 10-20. for small x because ofExpand
Uniform asymptotic expansions of the incomplete gamma functions and the incomplete beta function : (prepublication)
New asymptotic expansions are derived for the incomplete gamma functions and the incomplete beta function. In each case the expansion contains the complementary error function and an asymptoticExpand
The Percentile Points of Distributions Having Known Cumulants
In an earlier study of the uses of moments and cumulants in the specification of statistical distributions, the authors developed explicit asymptotic expansions, expressing any desired percentileExpand
The asymptotic expansion of the incomplete gamma functions : (preprint)
Earlier investigations on uniform asymptotic expansions of the incomplete gamma functions are reconsidered. The new results include estimations for the remainder and the extension of the results toExpand
Computation of the Percentage Points of the Chi-Square Distribution
Abstract : A Fortran IV program is developed for computing the percentage points of the Chi-square distribution. The Newton-Raphson procedure is used with good initial approximations. The maximumExpand
ACM Transactions on Mathematical Software
  • ACM Transactions on Mathematical Software
  • 1986