In this paper, we propose two bootstrap procedures, namely parametric and block bootstrap, to approximate the finite sample distribution of change-point estimators for piecewise stationary time series. The bootstrap procedures are then used to develop a generalized likelihood ratio scan method (GLRSM) for multiple change-point inference in piecewise stationary time series, which estimates the number and locations of change-points and provides a confidence interval for each change-point. The computational complexity of using GLRSM for multiple change-point detection is as low as $O(n(\log n)^{3})$ for a series of length n. Extensive simulation studies are provided to demonstrate the effectiveness of the proposed methodology under different scenarios. Applications to financial time series are also illustrated.

Severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) pandemic is still ongoing along with the global vaccination efforts against it. Here, we aimed to understand the longevity and strength of anti-SARS-CoV-2 IgG responses in a small community (n = 283) six months following local SARS-COV-2 outbreak in March 2020. Three serological assays were compared and neutralisation capability was also determined. Overall 16.6% (47/283) of the participants were seropositive and 89.4% (42/47) of the IgG positives had neutralising antibodies. Most of the symptomatic individuals confirmed as polymerase chain reaction (PCR) positive during the outbreak were seropositive (30/32, 93.8%) and 33.3% of the individuals who quarantined with a PCR confirmed patient had antibodies. Serological assays comparison revealed that Architect (Abbott) targeting the N protein LIASON® (DiaSorin) targeting the S protein and enzyme-linked immunosorbent assay (ELISA) targeting receptor binding domain detected 9.5% (27/283), 17.3% (49/283) and 17% (48/283), respectively, as IgG positives. The latter two assays highly agreed (kappa = 0.89) between each other. In addition, 95%, (19/20, by ELISA) and 90.9% (20/22, with LIASON) and only 71.4% (15/21, by Architect) of individuals that were seropositive in May 2020 were found positive also in September. The unexpected low rate of overall immunity indicates the absence of un-noticed, asymptomatic infections. Lack of overall high correlation between the assays is attributed mainly to target-mediated antibody responses and suggests that using a single serological assay may be misleading.

During the COVID-19 crisis, the French National Institute of Statistics and Economic Studies (INSEE) used aggregated and anonymous counting indicators based on network signaling data of three of the four mobile network operators (MNOs) in France to measure the distribution of population over the territory during and after the lockdown and to enrich the toolbox of high-frequency economic indicators used to follow the economic situation. INSEE’s strategy was to combine information coming from different MNOs together with the national population estimates it usually produces in order to get more reliable statistics and to measure uncertainty. This paper relates and situates this initiative within the long-term methodological collaborations between INSEE and different MNOs, and INSEE, Eurostat, and some other European national statistical institutes (NSIs). These collaborations aim at constructing experimental official statistics on the population present in a given place and at a given time, from mobile phone data (MPD). The COVID-19 initiative has confirmed that more methodological investments are needed to increase relevance of and trust in these data. We suggest this methodological work should be done in close collaboration between NSIs, MNOs, and research, to construct the most reliable statistical processes. This work requires exploiting raw data, so the research and statistical exemptions present in the general data protection regulation (GDPR) should be introduced as well in the new e-privacy regulation. We also raise the challenges of articulating commercial and public interest rationales and articulating transparency and commercial secrets requirements. Finally, it elaborates on the role NSIs can play in the MPD valorization ecosystem.

We assessed severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) reverse transcriptase-polymerase chain reaction (RT-PCR) diagnostic sensitivity and cycle threshold (Ct) values relative to symptom onset in symptomatic coronavirus disease-2019 (COVID-19) patients from Bavaria, Germany, of whom a subset was repeatedly tested. Locally weighted scatterplot smoothing method was used to assess the relationship between symptom onset and Ct-values. Kaplan−Meier plots were used to visualise the empirical probability of detecting viral ribonucleic acid (RNA) over time and estimate the time until clearance of viral RNA among the repeatedly tested patients. Among 721 reported COVID-19 cases, the viral RNA was detected in specimens taken between three days before and up to 48 days after symptom onset. The mean Ct-value was 28.6 (95% confidence interval (CI) 28.2–29.0) with the lowest mean Ct-value (26.2) observed two days after symptom onset. Up to 7 days after symptom onset, the diagnostic sensitivity of the RT-PCR among repeatedly sampled patients (n = 208) remained above 90% and decreased to 50% at day 12 (95% CI 10.5–21.5). Our data provide valuable estimates to optimise the timing of sampling of individuals for SARS-CoV-2 detection. A considerable proportion of specimens sampled before symptom onset had Ct-values comparable with Ct-values after symptom onset, suggesting the probability of presymptomatic transmission.

Telecommunications data are being explored by many countries as a new source of data that can be incorporated into their national statistical systems. In particular, “mobile positioning data” are increasingly being used to study population movements and population distributions. However, the legal, ethical, and technical complexities of working with this type of data often pose many barriers, which can prevent the data from being used at the times when it is most urgently needed. We demonstrate how having a robust public–private partnership framework, a privacy-preserving technical setup, and a communications strategy already in place, prior to an emergency, can enable governments to harness the advantages of telecommunications data at the times when it is most valuable. However, even once these foundations are in place, the challenges of competing priorities, managing expectations, and maintaining communication with data consumers during a pandemic mean that the potential of the data is not automatically translated into direct impact. This highlights the importance of sensitisation exercises, targeted at potential data users, to make clear the potential and limitations of the data, as well as the importance of being able to maintain direct communication with data users. The views expressed in this work belong solely to the authors and should not be interpreted as the views of their institutions.

In the absence of an established gold standard, an understanding of the testing cycle from individual exposure to test outcome report is required to guide the correct interpretation of severe acute respiratory syndrome-coronavirus-2 reverse transcriptase real-time polymerase chain reaction (RT-PCR) results and optimise the testing processes. Bayesian network models have been used within healthcare to bring clarity to complex problems. We use this modelling approach to construct a comprehensive framework for understanding the real-world predictive value of individual RT-PCR results.

We elicited knowledge from domain experts to describe the test process through a facilitated group workshop. A preliminary model was derived based on the elicited knowledge, then subsequently refined, parameterised and validated with a second workshop and one-on-one discussions.

Causal relationships elicited describe the interactions of pre-testing, specimen collection and laboratory procedures and RT-PCR platform factors, and their impact on the presence and quantity of virus and thus the test result and its interpretation. By setting the input variables as ‘evidence’ for a given subject and preliminary parameterisation, four scenarios were simulated to demonstrate potential uses of the model.

The core value of this model is a deep understanding of the total testing cycle, bridging the gap between a person's true infection status and their test outcome. This model can be adapted to different settings, testing modalities and pathogens, adding much needed nuance to the interpretations of results.

To investigate the current epidemiology of melioidosis in Yangon, Myanmar, between June 2017 and May 2019 we conducted enhanced surveillance for melioidosis in four tertiary hospitals in Yangon, where the disease was first discovered in 1911. Oxidase-positive Gram-negative rods were obtained from the microbiology laboratories and further analysed at the Department of Medical Research. Analysis included culture on Ashdown agar, the three disc sensitivity test (gentamicin, colistin and co-amoxiclav), latex agglutination, API 20 NE, antibiotic susceptibility testing, and a subset underwent molecular confirmation with a Burkholderia pseudomallei specific assay. Twenty one of 364 isolates (5.7%) were confirmed as B. pseudomallei and were mostly susceptible to the antibiotics used in standard therapy for melioidosis. Ten patients were from Yangon Region, nine were from Ayeyarwaddy region, and one each was from Kayin and Rakhine States. A history of soil contact was given by seven patients, five had diabetes mellitus and one had renal insufficiency. The patients presented with septicaemia (12 cases), pneumonia (three cases), urinary tract infection (two cases) and wound infection (four cases). Eighteen patients survived to hospital discharge. This study highlights the likelihood that melioidosis may be far more common, but underdiagnosed, in more rural parts of Myanmar as in other countries in SE Asia.

Developing agents capable of commonsense reasoning is an important goal in Artificial Intelligence (AI) research. Because commonsense is broadly defined, a computational theory that can formally categorize the various kinds of commonsense knowledge is critical for enabling fundamental research in this area. In a recent book, Gordon and Hobbs described such a categorization, argued to be reasonably complete. However, the theory’s reliability has not been independently evaluated through human annotator judgments. This paper describes such an experimental study, whereby annotations were elicited across a subset of eight foundational categories proposed in the original Gordon-Hobbs theory. We avoid bias by eliciting annotations on 200 sentences from a commonsense benchmark dataset independently developed by an external organization. The results show that, while humans agree on relatively concrete categories like time and space, they disagree on more abstract concepts. The implications of these findings are briefly discussed.

Given a finite set $A \subseteq \mathbb{R}^d$, points $a_1,a_2,\dotsc,a_{\ell} \in A$ form an $\ell$-hole in A if they are the vertices of a convex polytope, which contains no points of A in its interior. We construct arbitrarily large point sets in general position in $\mathbb{R}^d$ having no holes of size $O(4^dd\log d)$ or more. This improves the previously known upper bound of order $d^{d+o(d)}$ due to Valtr. The basic version of our construction uses a certain type of equidistributed point sets, originating from numerical analysis, known as (t,m,s)-nets or (t,s)-sequences, yielding a bound of $2^{7d}$. The better bound is obtained using a variant of (t,m,s)-nets, obeying a relaxed equidistribution condition.

A (not necessarily proper) vertex colouring of a graph has clustering c if every monochromatic component has at most c vertices. We prove that planar graphs with maximum degree $\Delta$ are 3-colourable with clustering $O(\Delta^2)$. The previous best bound was $O(\Delta^{37})$. This result for planar graphs generalises to graphs that can be drawn on a surface of bounded Euler genus with a bounded number of crossings per edge. We then prove that graphs with maximum degree $\Delta$ that exclude a fixed minor are 3-colourable with clustering $O(\Delta^5)$. The best previous bound for this result was exponential in $\Delta$.