Generalization of the order-restricted information criterion for multivariate normal linear models (pre-print )

The Akaike information criterion for model selection presupposes that the parameter space is not subject to order restrictions or inequality constraints.Anraku (1999) proposed a modified version of this criterion, called the order-restricted information criterion, for model selection in the one-way analysis of variance model when the population means are monotonic.We propose a generalization of this to the case when the population means may be restricted by a mixture of linear equality and inequality constraints.If the model has no inequality constraints, then the generalized order-restricted information criterion coincides with the Akaike information criterion.Thus, the former extends the applicability of the latter to model selection in multi-way analysis of variance models when some models may have inequality constraints while others may not. Simulation shows that the information criterion proposed in this paper performs well in selecting the correct model.

An Akaike-type information criterion for model selection under inequality constraints (pre-print)

The Akaike information criterion for model selection presupposes that the parameter space is not subject to order restrictions or inequality constraints. Anraku (1999) proposed a modified version of this criterion, called the order-restricted information criterion, for model selection in the one-way analysis of variance model when the population means are monotonic. We propose a generalization of this to the case when the population means may be restricted by a mixture of linear equality and inequality constraints. If the model has no inequality constraints, then the generalized order-restricted information criterion coincides with the Akaike information criterion. Thus, the former extends the applicability of the latter to model selection in multi-way analysis of variance models when some models may have inequality constraints while others may not. Simulation shows that the information criterion proposed in this paper performs well in selecting the correct model.

Bayesian approaches to the weighted kappa-like inter-rater agreement measures

Inter-rater agreement measures are used to estimate the degree of agreement between two or more assessors. When the agreement table is ordinal, different weight functions that incorporate row and column scores are used along with the agreement measures. The selection of row and column scores is effectual on the estimated degree of agreement. The weighted measures are prone to the anomalies frequently seen in agreement tables such as unbalanced table structures or grey zones due to the assessment behaviour of the raters. In this study, Bayesian approaches for the estimation of inter-rater agreement measures are proposed. The Bayesian approaches make it possible to include prior information on the assessment behaviour of the raters in the analysis and impose order restrictions on the row and column scores. In this way, we improve the accuracy of the agreement measures and mitigate the impact of the anomalies in the estimation of the strength of agreement between the raters. The elicitation of prior distributions is described theoretically and practically for the Bayesian estimation of five agreement measures with three different weights using an agreement table having two grey zones. A Monte Carlo simulation study is conducted to assess the classification accuracy of the Bayesian and classical approaches for the considered agreement measures for a given level of agreement. Recommendations for the selection of the highest performing agreement measure and weight combination are made in the breakdown of the table structure and sample size.

Task Scheduling for Subsea Flexible Pipes Decommissioning

Subsea flexible pipelines removal is subject to order restrictions, mostly caused by crossings. It is proposed to create a computational algorithm to design an optimal order of vessel intervention over a field. A real field was studied, and, from it, the mathematical base model was created upon graph theory, with great correlation with the minimum feedback arc set problem. Vessel movements were discretized and reduced to removal, reposition, and cut, leading to a state search. A-star algorithm was implemented to guide the search for the solution. Then, the complete algorithm was built, tested in a minimal environment, and finally applied to the real instance. To improve performance, a beam search filtering was envisioned, using seven ranking functions. Constructed model is suspected to be NP-hard, by correlation to minimum feedback arc set problem, leading to a large space search. Instances containing under 100 crossings were solved optimally, without needing any assistance. After implementing the heuristics and beam search, solution time was lowered by about 20 times, demonstrating the effectiveness of the technique. Also, ranking functions for pipe repositioning based on crossing count led to better results than crossing density. For cutting, an approximation based on feedback arc set was used. GreedyFAS was employed and gave satisfactory results. Bigger instances containing around 3000 crossings could not be solved optimally in a reasonable time, even with the heuristics. Improvements in A-star estimation function and bound the solution branches might lead to an optimal solution for these larger instances. Model proposed simplifies the operational order decisions and helps build the scheduling of operations. As it is based on state search, other aspects in logistics, vessel capacities and steps in decommissioning processes may be added, adjusting the neighboring weights and branching, keeping the same core.

Stringent Government Policies Are Associated With a Lower COVID-19 Spread Rate

The outbreak of COVID-19 has prompted a wide range of policy responses from governments around the world. In this study, we investigate the effect of governmental policies on the spread of the COVID-19 in a cross-country setting using the Oxford COVID-19 Government Response Stringency Index. We find that stringent government policies overall, and the following policies in particular, are associated with a lower spread rate of COVID-19 cases: workplace closing, restrictions on gatherings, close of public transport, stay-at-home order, restrictions on internal movement, and international travel controls; while school closing and public events cancellation are not associated with a lower COVID-19 spread. After including all policies into one single regression and examining their associations simultaneously with the virus spread, we find that the two policies stand out and remain to have a negative association with the COVID-19 spread: close of public transport and restrictions on international travel. Finally, we show that when countries are more oriented toward a tight culture, their governmental strict policies effect on the spread of COVID-19 becomes 1.5 – 3 times stronger than countries more toward a loose culture. Our findings suggest that the governments need to carefully implement policies to cope with the COVID-19 spread in their own social and cultural context.

Receipt

Tip decomposition is a crucial kernel for mining dense subgraphs in bipartite networks, with applications in spam detection, analysis of affiliation networks etc. It creates a hierarchy of vertex-induced subgraphs with varying densities determined by the participation of vertices in butterflies (2, 2-bicliques). To build the hierarchy, existing algorithms iteratively follow a delete-update (peeling) process: deleting vertices with the minimum number of butterflies and correspondingly updating the butterfly count of their 2-hop neighbors. The need to explore 2-hop neighborhood renders tip-decomposition computationally very expensive. Furthermore, the inherent sequentiality in peeling only minimum butterfly vertices makes derived parallel algorithms prone to heavy synchronization. In this paper, we propose a novel parallel tip-decomposition algorithm - REfine CoarsE-grained Independent Tasks (RECEIPT) that relaxes the peeling order restrictions by partitioning the vertices into multiple independent subsets that can be concurrently peeled. This enables RECEIPT to simultaneously achieve a high degree of parallelism and dramatic reduction in synchronizations. Further, RECEIPT employs a hybrid peeling strategy along with other optimizations that drastically reduce the amount of wedge exploration and execution time. We perform detailed experimental evaluation of RECEIPT on a shared-memory multicore server. It can process some of the largest publicly available bipartite datasets orders of magnitude faster than the state-of-the-art algorithms - achieving up to 1100× and 64× reduction in the number of thread synchronizations and traversed wedges, respectively. Using 36 threads, RECEIPT can provide up to 17.1× self-relative speedup.

Evaluating Multinomial Order Restrictions with Bridge Sampling

Hypotheses concerning the distribution of multinomial proportions typically entail exact equality constraints that can be evaluated using standard tests. Whenever researchers formulate inequality constrained hypotheses, however, they must rely on sampling-based methods that are relatively inefficient and computationally expensive. To address this problem we developed a bridge sampling routine that allows an efficient evaluation of multinomial inequality constraints. An empirical application showcases that bridge sampling outperforms current Bayesian methods, especially when relatively little posterior mass falls in the restricted parameter space. The method is extended to mixtures between equality and inequality constrained hypotheses.