Work

A shifted gamma–logistic transmuted distribution with properties and applications to process monitoring via CUSUM control charts

Year: 2026
Type: article
Abstract: We introduce the shifted gamma–logistic transmuted (SGLT) distribution, obtained by applying the quadratic rank transmutation map (QRTM) to a shifted, power-transformed logistic baseline that coincides with the power half-logistic law. Closed-form expressions are derived for the cdf, pdf, reliability and hazard functions, quantile function, moments, moment generating function and order statistics. The three parameters are estimated by maximum likelihood; the score equations were re-derived and checked against finite differences, and standard errors come from the observed information. A Monte Carlo study ( N = 1000 replications) shows that mean squared errors decrease with the sample size, but that the transmutation parameter is poorly determined for n ≤ 100. A two-sided likelihood-ratio CUSUM chart for SGLT observations is constructed, with the decision interval calibrated by simulation to an in-control average run length of 370. On two industrial datasets the SGLT attains the lowest AIC and BIC among the eight models of the original comparison (Tables 2 and 3), but its λ = 0 sub-model attains lower values still, the likelihood-ratio test does not reject λ = 0, and bootstrap-calibrated KS, CvM and AD tests reject the SGLT fit for the glass-fibre data while not rejecting it for the carbon-fibre data. The results delimit what the added transmutation parameter does and does not contribute.
Language: English
Open Access status: gold
Estimated APC paid: $2,690
OpenAlex ID: W7235018787
FWCI:
Cites: 37
Cited by:
Related to: 10