Asymptotic Inversion of Incomplete Gamma Functions

Abstract

Abstract. The normalized incomplete gamma functions P(a, x) and Q(a, x) are inverted for large values of the parameter a . That is, x-solutions of the equations P(a,x)=p, Q(a,x) = q, p e [0, 1], q = 1 -p, are considered, especially for large values of a . The approximations are obtained by using uniform asymptotic expansions of the incomplete gamma functions in which an error function is the dominant term. The inversion problem is started by inverting this error function term. Numerical results indicate that for obtaining an accuracy of four correct digits, the method can already be used for a = 2, although a is a large parameter. It is indicated that the method can be applied to other cumulative distribution functions.

Asymptotics and Numerics of Zeros of Polynomials That Are Related to Daubechies Wavelets

“…We will not discuss the role of other zeros in the plane in the neighborhood of the diagonals. A similar phenomenon is explained in Temme (1995). The point is that, although we started with integer values of N, the incomplete beta function in (2.1) and (2.7) can be interpreted with general complex values of N. When N is not an integer there are more y zeros, but those occur on other Riemann sheets of the multi-valued incomplete beta function.…”

EM-based joint estimation and detection for multiple antenna cognitive radios

“…The error in the channel and noise variance estimation can be quantified through the comparison with the conventional multiple antenna ED under noise variance mismatch of 1.8dB.The semianalytical evaluation of the proposed JED scheme after 5 iterations is also presented. By targeting specific α 2 , {ĥ (5) 2,p } andσ 2 (5) 2,v are estimated for 10 5 system parameters realizations, i.e., 10 5 trials, from (12) and (13) respectively. Then, these estimates are used to getγ opt (5) 2 andP opt (5) d,2 (γ opt (5) 2 ) of each trial from (28) and (29) respectively.…”

“…where Γ inv (c, w) is the inverse incomplete gamma function of Γ(c, y) [13]. Similarly, under the hypothesis…”

Cornish-Fisher technique for estimating detection probabilities for compound clutter distributions

“…It is useful to note that a Cornish-Fisher expansion [9] applied to ζ(N ) yields a representation as a series in inverse powers of N that is quite accurate even for small values of N [10]:…”