On the distributions of the differences between the estimated autocorrelations and partial autocorrelations at the same lag

Pukkila (1982) introduced a new portmanteau-type test of white noise. The test is based on the differences between the estimated autocorrelations and the corresponding partial autocorrelations. In the present paper the distributions of these differences are consider in the case of normal white noise. Approximations to these distributions are given. Goodness of fit of these theoretical distributions and the corresponding observed distributions is studied using simulated time series of various lengths.     

A Subset Selection Procedure for Selecting the Exponential Population Having the Longest Mean Lifetime When the Guarantee Times are the Same

ABSTRACT

Consider k(≥ 2) independent exponential populations Π₁, Π₂, …, Π _k , having the common unknown location parameter μ ∈ (?∞, ∞) (also called the guarantee time) and unknown scale parameters σ₁, σ₂, …σ _k , respectively (also called the remaining mean lifetimes after the completion of guarantee times), σ _i  > 0, i = 1, 2, …, k. Assume that the correct ordering between σ₁, σ₂, …, σ _k is not known apriori and let σ_[i], i = 1, 2, …, k, denote the ith smallest of σ _j s, so that σ_[1] ≤ σ_[2] ··· ≤ σ_[k]. Then Θ _i  = μ + σ _i is the mean lifetime of Π _i , i = 1, 2, …, k. Let Θ_[1] ≤ Θ_[2] ··· ≤ Θ_[k] denote the ranked values of the Θ _j s, so that Θ_[i] = μ + σ_[i], i = 1, 2, …, k, and let Π_(i) denote the unknown population associated with the ith smallest mean lifetime Θ_[i] = μ + σ_[i], i = 1, 2, …, k. Based on independent random samples from the k populations, we propose a selection procedure for the goal of selecting the population having the longest mean lifetime Θ_[k] (called the “best” population), under the subset selection formulation. Tables for the implementation of the proposed selection procedure are provided. It is established that the proposed subset selection procedure is monotone for a general k (≥ 2). For k = 2, we consider the loss measured by the size of the selected subset and establish that the proposed subset selection procedure is minimax among selection procedures that satisfy a certain probability requirement (called the P*-condition) for the inclusion of the best population in the selected subset.     

Using the Residual White Noise Autoregressive Order Determination Criterion to Identify Unit Roots in Arima Models

We present a simplified form of a univariate identification approach for time series models based on the residual white noise autoregressive order determination criterion and linear estimation methods. We also show how the procedure can be used to identify the degree of differencing necessary to induce stationarity in data. The performance of this approach is also contrasted with Portmanteau tests for detection of white noise residuals and with Dickey-Fuller and Bayesian procedures for detection of unit roots. Simulated and economic data are used to demonstrate the capabilities of the modified approach.     

Record values and related statistics - a review

In this brief survey, we discuss major developments of the past decade in the study of record values, record times, inter-record times and some related statistics from a series of observations     

Testing Goodness of Fit of Parametric AFT and PH Models with Residuals

In this article, we consider the problems of testing the goodness of fit of the parametric accelerated failure time model and the Cox proportional hazards model. We consider omnibus test statistics based on residuals. The statistical distributions of Kolmogorov, Cramer-von Mises–Smirnov, and Anderson–Darling statistics are all investigated by means of Monte Carlo simulations. Type-I, Type-II, and independent random censoring situations are all considered in this study. A Monte Carlo power study has also been carried out for these tests to distinguish between various baseline models, which reveals that the Anderson–Darling test performs better than the others.     

Fixed-width sequential confidence interval for the correlation coefficient of a bivariate normal distribution

A sequential confidence interval of fixed width 2d d > 0, is constructed for the correlation coefficient of a bivariate normal distribution. It is shown that the coverage probability is approximately equal to a preassigned number γ, 0 < γ < as d → 0.     

Work–Life Balance for Fathers in Globalized Knowledge Work. Some Insights from the Norwegian Context

This article takes as its point of departure the introduction of a new flexible time regime in parts of working life. There has been increased focus on how knowledge work in particular is developing into total commitment organizations where employees put in more and more time at work. Using two case studies from law and computing companies the article focuses on the organization of work and the organization of time in globalized knowledge organizations, and what effect this has on the time practices of male employees who are fathers. In the same period the Nordic countries have introduced state incentives and regulations that aim to help fathers achieve work–life balance. The article also discusses whether this is a fruitful policy, or whether fathers working in flexible time cultures need more deregulation, individual choice and flexible policy measures.     

The Crucial Role of Nomothetic and Idiographic Conceptions of Time: Interdisciplinary Collaboration in Nuclear Waste Management

The disposal of nuclear waste involves extensive time scales. Technical experts consider up to 1 million years for the disposal of spent fuel and high‐level waste in their safety assessment. Yet nuclear waste is not only a technical but also a so‐called sociotechnical problem and, therefore, requires interdisciplinary collaboration between technical, natural, social sciences, and the humanities in its management. Given that these disciplines differ in their language, epistemics, and interests, such collaboration might be problematic. Based on evidence from cognitive psychology, we suggest that, in particular, a concept like time is presumably critical and can be understood differently. This study explores how different scientific disciplines understand extensive time scales in general and then focuses on nuclear waste. Eighteen qualitative exploratory interviews were conducted with experts for time‐related phenomena of different disciplines, among them experts working in nuclear waste management. Analyses revealed two distinct conceptions of time corresponding to idiographic and nomothetic research approaches: scientists from the humanities and social sciences tend to have a more open, undetermined conception of time, whereas natural scientists tend to focus on a more determined conception that includes some undetermined aspects. Our analyses lead to reflections on potential difficulties for interdisciplinary teams in nuclear waste management. We focus on the understanding of the safety assessment, on potential implications for communication between experts from different disciplines (e.g., between experts from the humanities and engineering for risk assessment and risk communication), and we reflect on the roles of different disciplines in nuclear waste management.     

Selection procedures based on normal and nonnormal variables with single and multiple characters

The performance of selection procedures using a single screening variable are assessed in the presence of nonnormality, in particular mixtures of bivariate normal distributions and the bivariate Edgeworth series distribution.

Screening with multiple characters in the normal situation is studied using principal components.     

Model Identification in Dynamic Regression (Distributed Lag) Models

Dynamic regression models (also known as distributed lag models) are widely used in engineering for quality control and in economics for forecasting. In this article I propose a procedure for specifying such models in practice. The proposed procedure requires no prewhitening and can directly handle the nonstationary series. Furthermore, the procedure cross-validates prior beliefs about causal relationships between variables with empirical findings to ensure the suitability of model structure. An illustrative example is given.