Multiple Dependent State Sampling Inspection Plan for Lindley Distributed Quality Characteristic

This article develops multiple dependent state (MDS) sampling inspection plans based on the mean of lifetime quality characteristic that follows non-normal distributions viz., exponential and Lindley distribution. In this plan, the lot quality is measured by the lot mean (𝜇). We have estimated the optimal plan parameters of the proposed technique by non-linear optimization approaches considering acceptable quality level and rejection quality level. We have compared the sample size between the MDS sampling inspection plan and the single sampling inspection plan for the variable. Finally, we have taken two examples to illustrate the proposed technique.

Branching Process Modelling of COVID-19 Pandemic Including Immunity and Vaccination

We propose modeling COVID-19 infection dynamics using a class of two-type branching processes. These models require only observations on daily statistics to estimate the average number of secondary infections caused by a host and to predict the mean number of the non-observed infected individuals. The development of the epidemic process depends on the reproduction rate as well as on additional facets as immigration, adaptive immunity, and vaccination. Usually, in the existing deterministic and stochastic models, the officially reported and publicly available data are not sufficient for estimating model parameters. An important advantage of the proposed model, in addition to its simplicity, is the possibility of direct computation of its parameters estimates from the daily available data. We illustrate the proposed model and the corresponding data analysis with data from Bulgaria, however they are not limited to Bulgaria and can be applied to other countries subject to data availability.

Homogeneous Branching Processes with Non-Homogeneous Immigration

The stationary immigration has a limited effect over the asymptotic behavior of the underlying branching process. It affects mostly the limiting distribution and the life-period of the process. In contrast, if the immigration rate changes over time, then the asymptotic behavior of the process is significantly different and a variety of new phenomena are observed. In this review we discuss branching processes with time non-homogeneous immigration. Our goal is to help researchers interested in the topic to familiarize themselves with the current state of research.

Limit Theorems for a Strongly Supercritical Branching Process with Immigration in Random Environment

We consider a strongly supercritical branching process in random environment with immigration stopped at a distant time 𝑛. The offspring reproduction law in each generation is assumed to be geometric. The process is considered under the condition of its extinction after time 𝑛. Two limit theorems for this process are proved: the first one is for the time interval from 0 till 𝑛, and the second one is for the time interval from 𝑛 till + ∞ +\infty .

Asymptotic Properties of a Supercritical Branching Process with Immigration in a Random Environment

This is a short survey about asymptotic properties of a supercritical branching process ( Z n ) (Z_{n}) with immigration in a stationary and ergodic or independent and identically distributed random environment. We first present basic properties of the fundamental submartingale ( W n ) (W_{n}) , about the a.s. convergence, the non-degeneracy of its limit 𝑊, the convergence in L p L^{p} for p ≥ 1 p\geq 1 , and the boundedness of the harmonic moments E ⁢ W n - a \mathbb{E}W_{n}^{-a} , a > 0 a>0 . We then present limit theorems and large deviation results on log ⁡ Z n \log Z_{n} , including the law of large numbers, large and moderate deviation principles, the central limit theorem with Berry–Esseen’s bound, and Cramér’s large deviation expansion. Some key ideas of the proofs are also presented.

Critical Galton–Watson Processes with Overlapping Generations

A properly scaled critical Galton–Watson process converges to a continuous state critical branching process ξ ⁢ ( ⋅ ) \xi(\,{\cdot}\,) as the number of initial individuals tends to infinity. We extend this classical result by allowing for overlapping generations and considering a wide class of population counts. The main result of the paper establishes a convergence of the finite-dimensional distributions for a scaled vector of multiple population counts. The set of the limiting distributions is conveniently represented in terms of integrals ( ∫ 0 y ξ ⁢ ( y - u ) ⁢ d u γ \int_{0}^{y}\xi(y-u)\,du^{\gamma} , y ≥ 0 y\geq 0 ) with a pertinent γ ≥ 0 \gamma\geq 0 .

The SPRT Sign Chart for Process Dispersion

A nonparametric sequential probability ratio test control chart to monitor the process dispersion based on the sequential sign statistic is proposed. The statistical performance of this chart is evaluated by comparing it with that of the charts for dispersion based on sign statistic in the existing literature. It is found that the proposed chart outperforms all these charts uniformly in detecting a shift of any size over a wide range. An implementation of the chart is illustrated through an example.

Kernel and CDF-Based Estimation of Extropy and Entropy from Progressively Type-II Censoring with Application for Goodness of Fit Problems

Recently, entropy and extropy-based tests for the uniform distribution have attracted the attention of some researchers. This paper proposes nonparametric entropy and extropy estimators based on progressive type-II censoring and investigates their properties and behavior. Performance of the proposed estimators is studied via simulations. Entropy and extropy-based goodness-of-fit tests for uniformity are developed by the well performed estimators. The powers of the proposed uniformity tests are compared also via simulations assuming various alternatives and censoring schemes.