In general the exact distribution of a convolution of independent gamma random variables is quite complicated and does not admit a closed form. Of all the distributions proposed, the gamma-series representation of Moschopoulos (1985) is relatively simple to implement but for particular combinations of scale and/or shape parameters the computation of the weights of the series can result complicated and too much time consuming to allow a large-scale application. Recently a compact random parameter representation of the convolution has been proposed by Vellaisamy and Upadhye (2009) and it allows to give an exact interpretation to the weights of the series. They describe an infinite discrete probability distribution. This result suggested to approximate Moschopoulos’s expression looking for an approximating theoretical discrete distribution for the weights of the series. More precisely, we propose a general negative binomial distribution. The result is an ”excellent” approximation, fast and simple to implement for any parameter combination.

An approximation to the convolution of Gamma Distributions / Barnabani, Marco. - In: COMMUNICATIONS IN STATISTICS. SIMULATION AND COMPUTATION. - ISSN 0361-0918. - STAMPA. - .......:(2014), pp. 0-0. [10.1080/03610918.2014.963612]

An approximation to the convolution of Gamma Distributions

BARNABANI, MARCO
2014

Abstract

In general the exact distribution of a convolution of independent gamma random variables is quite complicated and does not admit a closed form. Of all the distributions proposed, the gamma-series representation of Moschopoulos (1985) is relatively simple to implement but for particular combinations of scale and/or shape parameters the computation of the weights of the series can result complicated and too much time consuming to allow a large-scale application. Recently a compact random parameter representation of the convolution has been proposed by Vellaisamy and Upadhye (2009) and it allows to give an exact interpretation to the weights of the series. They describe an infinite discrete probability distribution. This result suggested to approximate Moschopoulos’s expression looking for an approximating theoretical discrete distribution for the weights of the series. More precisely, we propose a general negative binomial distribution. The result is an ”excellent” approximation, fast and simple to implement for any parameter combination.
2014
.......
0
0
Barnabani, Marco
File in questo prodotto:
File Dimensione Formato  
SumGammaReviewed.pdf

Accesso chiuso

Tipologia: Pdf editoriale (Version of record)
Licenza: Open Access
Dimensione 120.21 kB
Formato Adobe PDF
120.21 kB Adobe PDF   Richiedi una copia

I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificatore per citare o creare un link a questa risorsa: https://hdl.handle.net/2158/900533
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 3
  • ???jsp.display-item.citation.isi??? 3
social impact