Search

In this study, in order to do a life testing experiment, sampled units are divided into a prefixed number of groups with equal number of units. Units in all the groups are tested simultaneously and independently and, in each group the experiment is terminated as soon as a prefixed time elapses or a prefixed number of failures occurs. We provide the meta-analysis of an exponential lifetime data from Type-I hybrid censored samples. The main goal of this study is to obtain optimal schemes based on some optimality criteria by minimizing certain cost function that is based on a maximum likelihood estimator of mean lifetime. We provide the maximum likelihood estimator of mean lifetime and its probability density function under this set-up. Various optimal schemes have been provided by minimizing expected total cost incurred during the experiment as the raw moments can be obtained explicitly. Numerical results on bias and mean squared error of the maximum likelihood estimator have been reported. We also provide confidence intervals of the unknown parameter. For illustration, meta-analysis for a real data set of three groups is presented. [ABSTRACT FROM AUTHOR]

In this article, we provide a consistent method of estimation for the parameters of a three-parameter generalized exponential distribution which avoids the problem of unbounded likelihood function. The method is based on a maximum likelihood estimation of the shape parameter, which uses location and scale invariant statistic, originally proposed by Nagatsuka et al. (A consistent method of estimation for the three-parameter weibull distribution, Computational Statistics & Data Analysis 58:210–26). It has been shown that the estimators are unique and consistent for the entire range of the parameter space. We also present a Monte-Carlo simulation study along with the comparisons with some prominent estimation methods in terms of the bias and root mean square error. For the illustration purpose, the data analysis of a real lifetime data set has been reported. [ABSTRACT FROM AUTHOR]

There is a substantial amount of literature in the area of acceptance sampling plan with censored lifetime data. However, the optimality of a Bayesian sampling plan in the presence of competing risks has not been considered so far. In this paper, first, the Bayesian sampling plans (BSP) for Type-II and Type-I hybrid censoring schemes are discussed in presence of competing risks when the lifetime distribution is exponential. The closed-form expression of the Bayes decision function is obtained analytically for a linear loss function. Then we consider the Weibull distribution with an unknown shape parameter under Type-I hybrid censoring scheme in presence of competing risks to obtain the BSP. However, the Bayes decision function cannot be obtained in closed-form for a general loss function, and in such cases, a numerical algorithm is proposed. As an illustration, in the exponential case, a quadratic loss function, and in Weibull case, a non-polynomial loss function, are considered for the application of the proposed numerical approach to obtain the optimum BSPs using the Bayes decision function. [ABSTRACT FROM AUTHOR]

In this paper, a multiple step-stress model is designed and analyzed when the data are Type-I censored. Lifetime distributions of the experimental units at each stress level are assumed to follow a two-parameter Weibull distribution. Further, distributions under each of the stress levels are connected through a tampered failure-rate based model. In a step-stress experiment, as the stress level increases, the load on the experimental units increases and hence the mean lifetime is expected to be shortened. Taking this into account, the aim of this paper is to develop the order restricted inference of the model parameters of a multiple step-stress model based on the frequentist approach. An extensive simulation study has been carried out and two real data sets have been analyzed for illustrative purposes. [ABSTRACT FROM AUTHOR]

In this paper, we consider the order estimation problem of a 2-dimensional complex superimposed exponential signal model in presence of additive white noise. We use the recently proposed exponentially embedded family (EEF) rule (see Stoica and Babu in IEEE Signal Process Lett 19(9):551–554, 2012; Kay in IEEE Trans Aerosp Electron Syst 41(1):333–345, 2005) for estimating the order of the 2-dimensional signal model and prove that the EEF rule based estimator is consistent in large sample scenario. Extensive simulations are performed to ascertain the performance of the order estimation rule and also to compare the finite sample performance of EEF rule based estimator with other popular order selection rules using simulation examples. [ABSTRACT FROM AUTHOR]

Accurate estimation of the parameters of superimposed sinusoidal signals is an important problem in digital signal processing and time series analysis. In this article, we propose a simultaneous estimation procedure for estimation of the number of signals and signal parameters. The proposed sequential method is based on a robust bivariate M-periodogram and uses the orthogonal structure of the superimposed sinusoidal model for sequential estimation. Extensive simulations and data analysis show that the proposed method has a high degree of frequency resolution capability and can provide robust and efficient estimates of the number of signals and signal parameters. [ABSTRACT FROM AUTHOR]

In this paper, we consider the problem of estimating the number of components of a superimposed nonlinear sinusoids model of a signal in the presence of additive noise. We propose and provide a detailed empirical comparison of robust methods for estimation of the number of components. The proposed methods, which are robust modifications of the commonly used information theoretic criteria, are based on various M-estimator approaches and are robust with respect to outliers present in the data and heavy-tailed noise. The proposed methods are compared with the usual non-robust methods through extensive simulations under varied model scenarios. We also present real signal analysis of two speech signals to show the usefulness of the proposed methodology. [ABSTRACT FROM AUTHOR]

Accurate and timely forecasting of future business cycle movements is of high priority for well comprehensible reasons, where the organised business sector takes a pivotal role in the well-being of the economy in both developing and developed nations. The original series to be modeled is first decomposed into various scale related wavelet components using wavelet-filtering technique. The decomposed components, represent components of the original rime series, like, the long term smooth component, the seasonal and the cyclical components, at different periodicities. Comparison of results of the proposed model with usual genetically optimized neural network model, standard statistical rime series models, time series and neural network models based on Fourier representation, indicates superior performance of the proposed method. [ABSTRACT FROM AUTHOR]