Journal of Data Science

Abstract: In this article, we studied three types of time series analysis methods in modeling and forecasting the severe acute respiratory syndrome (SARS) epidemic in mainland China. The first model was a Box-Jenkins model, autoregressive model with order 1 (AR(1)). The second model was a random walk (ARIMA(0,1,0)) model on the log transformed daily reported SARS cases and the third one was a combination of growth curve fitting and autoregressive moving average model, ARMA(1,1). We applied all these three methods to monitor the dynamic of SARS in China based on the daily probable new cases reported by the Ministry of Health of China.

In this paper Zografos Balakrishnan Power Lindley (ZB-PL) distribution has been obtained through the generalization of Power Lindley distribution using Zografos and Balakrishnan (2009) technique. For this technique, density of upper record values exists as their special case. Probability density (pdf), cumulative distribution (cdf) and hazard rate function (hrf) of the proposed distribution are obtained. The probability density and cumulative distribution function are expanded as linear combination of the density and distribution function of Exponentiated Power Lindley (EPL) distribution. This expansion is further used to study different properties of the new distribution. Some mathematical and statistical properties such as asymptotes, quantile function, moments, mgf, mean deviation, renyi entropy and reliability are also discussed. Probability density (pdf), cumulative distribution (cdf) and hazard rate (hrf) functions are graphically presented for different values of the parameters. In the end Maximum Likelihood Method is used to estimate the unknown parameters and application to a real data set is provided a. It has been observed that the proposed distribution provides superior fit than many useful distributions for given data set.

In this article, we introduce a new class of five-parameter model called the Exponentiated Weibull Lomax arising from the Exponentiated Weibull generated family. The new class contains some existing distributions as well as some new models. Explicit expressions for its moments, distribution and density functions, moments of residual life function are derived. Furthermore, Rényi and q–entropies, probability weighted moments, and order statistics are obtained. Three suggested procedures of estimation, namely, the maximum likelihood, least squares and weigthed least squares are used to obtain the point estimators of the model parameters. Simulation study is performed to compare the performance of different estimates in terms of their relative biases and standard errors. In addition, an application to two real data sets demonstrate the usefulness of the new model comparing with some new models.

Abstract: The modified autoregressive (mAR) index has been proposed as a description of the clustering of shots of similar duration in a motion picture. In this paper we derive robust estimates of the mAR index for high grossing films at the US box office using a rank-based autocorrelation function resis tant to the influence of outliers and compare this to estimates obtained using the classical, moment-based autocorrelation function. The results show that (1) The classical mAR index underestimates both the level of shot clustering in a film and the variation in style among the films in the sample; (2) there is a decline in shot clustering from 1935 to the 1950s followed by an increase from the 1960s to the 1980s and a levelling off thereafter rather than the monotonic trend indicated by the classical index, and this is mirrored in the trend of the median shot lengths and interquartile range; and (3) the rank mAR index identifies differences between genres overlooked when using the classical index.

Abstract: In this note a new method of comparing component structural importance is introduced and compared to other existing ones. Especially, relationships of the new comparison method to the H-importance due to Hwang (2001,2005), the criticality ordering due to Boland et al. (1989) and Birnbaum importance are obtained. Illustrative examples are given.

In semiparametric regression it is of interest to detect anomalous observations that exert an unduly large influence on the parameter’s esti-mate and fitted values. Usually the existence of influential observations is complicated by the presence of collinearity. However no method of influ-ence diagnostics available for the possible effects that collinearity can have on the influence of an observation on the estimates of parametric and non-parametric component of semiparametric regression models. In this paper we show when Liu estimators are used to mitigate the effects of collinearity the influence of some observations can be drastically modified. We propose a case deletion formula to detect influential points in Liu estimators of semi-parametric regression models . As an illustrative example a real data set are analysed.

Abstract: The traditional method for processing functional magnetic resonance imaging (FMRI) data is based on a voxel-wise, general linear model. For experiments conducted using a block design, where periods of activation are interspersed with periods of rest, a haemodynamic response function (HRF) is convolved with the design function and, for each voxel, the convolution is regressed on prewhitened data. An initial analysis of the data often involves computing voxel-wise two-sample t-tests, which avoids a direct specification of the HRF. Assuming only the length of the haemodynamic delay is known, scans acquired in transition periods between activation and rest are omitted, and the two-sample t-test is used to compare mean levels during activation versus mean levels during rest. However, the validity of the two-sample t-test is based on the assumption that the data are Gaussian with equal variances. In this article, we consider the Wilcoxon rank test as well as modified versions of the classical t-test that correct for departures from these assumptions. The relative performance of the tests are assessed by applying them to simulated data and comparing their size and power; one of the modified tests (the CW test) is shown to be superior.

Abstract: Accelerated life testing (ALT) has gained greater importance because of dealing with high reliability units. As a result, there is a big need to use a goodness of fit (GOF) technique for testing the underlying lifetime distribution. But there is a difficulty due to the existence of several stress levels with different samples of units at each level. Then, the choice of a certain GOF technique is based on its capability to combine the failure times from all stress levels to reach a conclusion about the adequacy of a certain lifetime distribution at each stress level. In this paper, the extended Neyman’s smooth test (ENST) is chosen. It is then modified in order to be used in validating the distributional assumption of accelerated failure time (AFT) model. This modified method is called; the adapted extended Neyman’s smooth test (AENST). It is applied to test for both Weibull and exponential distributions in case of constant stress under complete sampling. To check the performance of the AENST, a comparison is made with the conditional probability integral transformation test (CPITT) via a simulation study. Moreover, a real data set is provided to illustrate the application of the introduced AENST. The results revealed that the AENST is a powerful test comparing with the CPITT. Thus, the AENST is recommended for testing the AFT models.