Inference for a truncated positive normal distribution
- 66 Downloads
The main object of this paper is to study an extension of the half normal distribution defined by adding a positive truncation to it. The new model is more flexible than the half-normal distribution and contains the half normal distribution as a special case. Properties of this distribution, such as moments, hazard function and entropy are studied and parameters estimation is dealt with by using moments and maximum likelihood. A real data application indicates good fit performance of the new model when compared to other competitors in literatures.
Keywordshalf-normal distribution maximum likelihood truncation
MR Subject Classification62E15 62F10
Unable to display preview. Download preview PDF.
The authors acknowledge helpful comments and suggestions by the referee which substantially improved the presentation. The research of HJGómez, NMOlmos and HVarela was supported by SEMILLERO-UA2014 (Chile). The research of HBolfarine was supported by CNPq and Fapesp (Brasil).
- C E Bonferroni. Elmenti di statistica generale, Libreria Seber, Firenze, 1930.Google Scholar
- C Y Chou, H R Liu. Properties of the half-normal distribution and its application to quality control, J Ind Technol, 1998, 14(13): 4–7.Google Scholar
- N L Jonhson, S Kotz, N Balakrishnan. Continuous Univariate Distributions, Vol 2, 2nd ed, Wiley, New York, 1995.Google Scholar