Keywords: INAR(1) model Negative binomial thinning Bivariate integer-valued time series model a b s t r a c t In this paper we introduce a simple bivariate integer-valued time series model with positively correlated geometric marginals based on the negative binomial thinning mechanism. Some properties of the model are considered. The unknown parameters of… (More)
Mössbauer, FTIR and XRD analyses showed that in aqueous medium in air in the presence of L-tryptophan (Trp) or indole-3-acetic acid (IAA) the ambient-temperature ageing of the precipitates formed from ferrous sulphate at pH~7 gave composite phases with varying proportions of γ-FeOOH (a dominating crystalline phase), α-FeOOH (both fine-grained, showing… (More)
Porous silicon (PSi) samples were prepared by electrochemical anodisation of silicon on insulator layers. Structural and optical properties of prepared samples were investigated by Raman and photoluminescence (PL) spectroscopy and field emission scanning electron microscopy (FE-SEM). The anodisation of silicon on insulator layers was performed by… (More)
In this paper, we introduce a new (combined) integer-valued autoregressive model of order p with geometric marginal distributions, denoted by CGINAR(p), using the negative binomial thinning introduced by Ristić et al. . Several properties of the model are constructed and discussed, including one-step ahead conditional statistical measures. Some methods… (More)
Many if not most lifetime distributions are motivated only by mathematical interest. Here, a new three parameter distribution motivated mainly by lifetime issues is introduced. Some properties of the new distribution including estimation procedures are derived. Two real data applications are described to show superior performance versus at least seven of… (More)
Lewis and McKenzie have described the maximum process with marginal distribution U(, ∞). In this paper, we discuss some properties of this process. We also apply some estimation methods for estimating the parameter of the process. It is shown that the conditional least squares estimator is strongly consistent and asymptotically normal.