Search

Can we elaborate a traffic-sensitive eco-driving or GLOSA (Green Light Optimal Speed Advice) strategy with a frugal amount of data when approaching an intersection? Here is the purpose of this work, which aims to adapt a traffic-theory-based estimation of the expected queue-length within mixed traffic (Connected and non-Connected Vehicles) in the vicinity of a signalized intersection. While the expected queue-length methodology was developed recently and fits natively with Eulerian traffic indicators resulting from loop sensors or cameras, this paper adapts such a methodology to Lagrangian indicators as the traces produced by any Connected Vehicle, including Floating Car or Probe Data. The main interest of the methodology lies in the frugal amount of data and expenses required to perform the traffic-sensitive speed-advisory at any connected road intersection. The full methodology is developed to extend the SPAT messages broadcast to end-users and take advantage of the Cooperative Awareness Messages (CAM) acting as GPS traces for Connected Vehicles. Contrary to Eulerian-based indicators, no supplementary and costly investment is required to collect the input data and compute the queue-length estimation. However, applying strategies based on Lagrangian indicators will affect the direct traffic observation through these indicators. Therefore, it requires to develop an assessment and predictive framework to estimate the traffic conditions. The performance of the introduced methodology is compared to alternative methods, among other Eulerian-based methods. It results from the analysis that the introduced approach performs almost as well as the ones based on exhaustive, but costly data collections. [ABSTRACT FROM AUTHOR]

Background: Japan implemented strict border control measures and all incoming passengers were subject to entry screening with reverse transcription–polymerase chain reaction or antigen testing. From late 2020, exit screening within 72 h of departure to Japan also became mandatory. In this study, we evaluated the effectiveness of the exit screening policy in Japan by analyzing airport screening data from October 2020 to April 2022. Methods: In addition to assessing entry screening data over time of passengers from the United Kingdom, we examined the prevalence of coronavirus disease 2019 (COVID-19) in the United Kingdom based on the Office of National Statistics infection survey. We constructed a statistical model that described entry screening positivity over time using Office of National Statistics prevalence data as the explanatory variable. Ideally, the time-dependent patterns of entry screening and Office of National Statistics prevalence data should resemble each other; however, we found that, sometimes, they were different and regarded the difference to statistically partly reflect the effectiveness of exit screening. Results: The average proportion positive in one month before mandatory exit screening was implemented among Japanese passengers was 0.67% (95% confidence interval [CI]: 0.45, 0.98), whereas the proportion positive decreased to 0.49% (95% CI: 0.21, 1.15) in the first month of exit screening. Adjusting for time-dependent prevalence at the origin, we concluded that exit screening contributed to reducing passenger positivity by 59.3% (95% CI: 19.6, 81.3). The overall positivity values among passengers during the Delta and Omicron variant periods were 3.46 times and 1.46 times that during the pre-Delta variant period, respectively. Conclusions: We used a simplistic statistical model and empirical data from passengers arriving in Japan from the United Kingdom to support that exit screening helped to reduce the proportion positive by 59%. Although the proportion positive later increased considerably and precluded preventing the introduction of imported cases, submitting a certificate for a negative test result contributed to reducing the positivity among travelers. [ABSTRACT FROM AUTHOR]

Abstract Background Japan implemented strict border control measures and all incoming passengers were subject to entry screening with reverse transcription–polymerase chain reaction or antigen testing. From late 2020, exit screening within 72 h of departure to Japan also became mandatory. In this study, we evaluated the effectiveness of the exit screening policy in Japan by analyzing airport screening data from October 2020 to April 2022. Methods In addition to assessing entry screening data over time of passengers from the United Kingdom, we examined the prevalence of coronavirus disease 2019 (COVID-19) in the United Kingdom based on the Office of National Statistics infection survey. We constructed a statistical model that described entry screening positivity over time using Office of National Statistics prevalence data as the explanatory variable. Ideally, the time-dependent patterns of entry screening and Office of National Statistics prevalence data should resemble each other; however, we found that, sometimes, they were different and regarded the difference to statistically partly reflect the effectiveness of exit screening. Results The average proportion positive in one month before mandatory exit screening was implemented among Japanese passengers was 0.67% (95% confidence interval [CI]: 0.45, 0.98), whereas the proportion positive decreased to 0.49% (95% CI: 0.21, 1.15) in the first month of exit screening. Adjusting for time-dependent prevalence at the origin, we concluded that exit screening contributed to reducing passenger positivity by 59.3% (95% CI: 19.6, 81.3). The overall positivity values among passengers during the Delta and Omicron variant periods were 3.46 times and 1.46 times that during the pre-Delta variant period, respectively. Conclusions We used a simplistic statistical model and empirical data from passengers arriving in Japan from the United Kingdom to support that exit screening helped to reduce the proportion positive by 59%. Although the proportion positive later increased considerably and precluded preventing the introduction of imported cases, submitting a certificate for a negative test result contributed to reducing the positivity among travelers.

A single-photon avalanche diode (SPAD) is a photon-counting sensor renowned for its exceptional single-photon sensitivity. One significant feature of SPADs is their non-linear response to light, making them ideal for high-dynamic range imaging applications. In SPAD imaging, the photon detection mode, which depends on the quenching method employed, is crucial for optimizing image quality and dynamic range. This paper examines the free-running and gating imaging modes, evaluating their impacts on photon capture and saturation limits. Given that the number of incident photons follows a Poisson distribution, we introduce an innovative imaging-quenching model based on statistical mathematics. We designed and fabricated two SPAD imaging sensors using 180 nm CMOS technology. Image processing and evaluation were conducted using a mapping method. Our results show that in low-light conditions, the gating mode surpasses the free-running mode in the signal-to-noise ratio (SNR). However, the free-running mode exhibits a saturation limit of more than an order of magnitude higher than that of the gating mode, demonstrating its superior capability to handle a broader range of light intensities. This paper provides a thorough analysis of the differences between the two imaging methods, incorporating the theoretical mathematical model, circuit characteristics, and computed imaging quality. [ABSTRACT FROM AUTHOR]

Recently, interest in functional safety has surged because vehicle technology increasingly relies on electronics and automation. Failure of certain system components can endanger driver safety and is costly to address. The detection of abnormal data is crucial for enhancing the reliability, safety, and efficiency. This study introduces a novel anomaly-detection method of designable generative adversarial network anomaly detection (DGANomaly). DGANomaly combines the data augmentation method of a designable generative adversarial network (DGAN) with a generative adversarial network anomaly-detection data classification technique. DGANomaly not only generates virtual data that are challenging to obtain or simulate but also produces a range of statistical design variables for normal and abnormal data. This approach enables the specific identification of normal and abnormal design variables. To demonstrate its effectiveness, the DGANomaly method was applied to an electric power steering (EPS) model when multiple degradations of gear stiffness, gear friction, and rack displacement were considered. An EPS model was constructed and validated using simulation programs such as Prescan, Amesim, and Simulink. Consequently, DGANomaly exhibited a higher classification accuracy than the other methods, allowing for more accurate detection of abnormal data. Additionally, a clearer range of statistical designs can be obtained for normal data. These results indicate that the statistical design variables that are less likely to fail can be obtained using minimal data. [ABSTRACT FROM AUTHOR]

We give a thorough description of the asymptotic property of the maximum likelihood estimator (MLE) of the skewness parameter of a Skew Brownian Motion (SBM). Thanks to recent results on the Central Limit Theorem of the rate of convergence of estimators for the SBM, we prove a conjecture left open that the MLE has asymptotically a mixed normal distribution involving the local time with a rate of convergence of order 1/4. We also give a series expansion of the MLE and study the asymptotic behavior of the score and its derivatives, as well as their variation with the skewness parameter. In particular, we exhibit a specific behavior when the SBM is actually a Brownian motion, and quantify the explosion of the coefficients of the expansion when the skewness parameter is close to −1 or 1. [ABSTRACT FROM AUTHOR]

Abstract Background The Japanese government has instituted border control measures against COVID-19, including entry and exit screening of people arriving from overseas. We sought to evaluate the effectiveness of the exit screening policy in Japan in reducing the risk of importing COVID-19 cases among travelers from Asian and Pacific countries. Methods The study period was stratified based on the timing of exit screening: (i) the control period (the pre-exit screening period from 25 October 2020 to 16 January 2021), (ii) the time period with the Alpha variant from 17 January to 10 April 2021, and (iii) the time period with the Delta variant from 2 May to 2 October 2021. Incidence data in the countries of origin were used to adjust for the risk of infection among travelers. The positivity rate of entry screening in Japan was compared among the three different study periods, adjusting for the risk of infection in the country of origin. Results The adjusted relative risk of positivity was greatly reduced and substantially below the value of 1 during the Alpha variant period compared with the control period. Although the relative risks increased when comparing the Delta variant period against control, the estimate remained below 1, except for among travelers from India and Myanmar. The relative risk reduction was greatest in high-income countries, with estimates of 100% and 96% risk reduction during the Alpha and Delta variant periods, respectively, followed by upper-middle-income countries with estimates of 90% and 76%, respectively. Conclusions Even in the presence of the Alpha and Delta variants, exit screening clearly reduced the risk of infection among travelers arriving from Asian and Pacific nations. As the testing relies on the country of origin, the effectiveness varied greatly by the socioeconomic income status and epidemiological situation of those countries. Test standardization and quality assurance may be required in low- and middle-income countries.

Background: The Japanese government has instituted border control measures against COVID-19, including entry and exit screening of people arriving from overseas. We sought to evaluate the effectiveness of the exit screening policy in Japan in reducing the risk of importing COVID-19 cases among travelers from Asian and Pacific countries. Methods: The study period was stratified based on the timing of exit screening: (i) the control period (the pre-exit screening period from 25 October 2020 to 16 January 2021), (ii) the time period with the Alpha variant from 17 January to 10 April 2021, and (iii) the time period with the Delta variant from 2 May to 2 October 2021. Incidence data in the countries of origin were used to adjust for the risk of infection among travelers. The positivity rate of entry screening in Japan was compared among the three different study periods, adjusting for the risk of infection in the country of origin. Results: The adjusted relative risk of positivity was greatly reduced and substantially below the value of 1 during the Alpha variant period compared with the control period. Although the relative risks increased when comparing the Delta variant period against control, the estimate remained below 1, except for among travelers from India and Myanmar. The relative risk reduction was greatest in high-income countries, with estimates of 100% and 96% risk reduction during the Alpha and Delta variant periods, respectively, followed by upper-middle-income countries with estimates of 90% and 76%, respectively. Conclusions: Even in the presence of the Alpha and Delta variants, exit screening clearly reduced the risk of infection among travelers arriving from Asian and Pacific nations. As the testing relies on the country of origin, the effectiveness varied greatly by the socioeconomic income status and epidemiological situation of those countries. Test standardization and quality assurance may be required in low- and middle-income countries. [ABSTRACT FROM AUTHOR]

We consider two parsimonious models of binary high-order Markov chains and discover their ability to approximate arbitrary high-order Markov chains. Two types of global measures for approximation accuracy are introduced, theoretical and experimental results are obtained for these measures and for the considered parsimonious models. New consistent statistical parameter estimator is constructed for parsimonious model based on two-layer artificial neural network. [ABSTRACT FROM AUTHOR]

Many applied linguists have attempted to quantify the number of English-speaking populations, both globally and in specific countries. However, thus far, those in East Asia have not been estimated through a robust technique. This paucity of estimation can be partly attributed to the lack of censuses or large-scale sociolinguistic surveys; nonetheless, other secondary data can be used for this purpose. This study analyses two nationally representative surveys conducted in China, Japan, South Korea, and Taiwan – the East Asian Social Survey (conducted in 2008 and 2018) and AsiaBarometer (2003–2008) – and estimates their English-speaking population. The results indicate that English speakers (those who evaluated themselves as having a fair or high level of English proficiency) account for approximately 10% of the population in China and Japan and approximately 20% in South Korea and Taiwan. Furthermore, English proficiency varied according to generation and educational level in all four jurisdictions. Based on these findings, this article discusses cross-national variations of English-speaking populations in the four jurisdictions, the absolute number of speakers, its implications, and possible self-report biases and their assessment. [ABSTRACT FROM AUTHOR]

At the end of 2022, a total of 20,003 diagnoses of human immunodeficiency virus (HIV) infection and 8,983 cases of acquired immunodeficiency syndrome (AIDS) among Japanese nationals, and 3,860 HIV diagnoses and 1,575 AIDS cases among foreign residents, had been notified to the government in Japan. This study updates the estimate of HIV incidence, including during the COVID-19 pandemic. It aimed to reconstruct the incidence of HIV and understand how the disruption caused by COVID-19 affected the epidemiology of HIV. Using a median incubation period of 10.0 years, the number of undiagnosed HIV infections was estimated to be 3,209 (95% confidence interval (CI): 2,642, 3,710) at the end of 2022. This figure has declined steadily over the past 10 years. Assuming that the median incubation period was 10.0 years, the proportion of diagnosed HIV infections, including surviving AIDS cases, was 89.3% (95% CI: 87.8%, 91.0%). When AIDS cases were excluded, the proportion was 86.2% (95% CI: 84.3%, 88.3%). During the COVID-19 pandemic, the estimated annual diagnosis rate was slightly lower than during earlier time intervals, at around 16.5% (95% CI: 14.9%, 18.1%). Japan may already have achieved diagnostic coverage of 90%, given its 9% increment in the diagnosed proportion during the past 5 years. The incidence of HIV infection continued to decrease even during the COVID-19 pandemic from 2020 to 2022, and the annual rate of diagnosis decreased slightly to 16.5%. Monitoring the recovery of diagnosis along with the effective reproduction number is vital in the future.

Abstract Background A major epidemic of COVID-19 caused by the Delta variant (B.1.617.2) occurred in India from March to July 2021, resulting in 19 million documented cases. Given the limited healthcare and testing capacities, the actual number of infections is likely to have been greater than reported, and several modelling studies and excess mortality research indicate that this epidemic involved substantial morbidity and mortality. Methods To estimate the incidence during this epidemic, we used border entry screening data in Japan to estimate the daily incidence and cumulative incidence of COVID-19 infection in India. Analysing the results of mandatory testing among non-Japanese passengers entering Japan from India, we calculated the prevalence and then backcalculated the incidence in India from February 28 to July 3, 2021. Results The estimated number of infections ranged from 448 to 576 million people, indicating that 31.8% (95% confidence interval (CI): 26.1, 37.7) – 40.9% (95% CI: 33.5, 48.4) of the population in India had experienced COVID-19 infection from February 28 to July 3, 2021. In addition to obtaining cumulative incidence that was consistent with published estimates, we showed that the actual incidence of COVID-19 infection during the 2021 epidemic in India was approximately 30 times greater than that based on documented cases, giving a crude infection fatality risk of 0.47%. Adjusting for test-negative certificate before departure, the quality control of which was partly questionable, the cumulative incidence can potentially be up to 2.3–2.6 times greater than abovementioned estimates. Conclusions Our estimate of approximately 32–41% cumulative infection risk from February 28 to July 3, 2021 is roughly consistent with other published estimates, and they can potentially be greater, given an exit screening before departure. The present study results suggest the potential utility of border entry screening data to backcalculate the incidence in countries with limited surveillance capacity owing to a major surge in infections.

Accurate measurement of daily infection incidence is crucial to epidemic response. However, delays in symptom onset, testing, and reporting obscure the dynamics of transmission, necessitating methods to remove the effects of stochastic delays from observed data. Existing estimators can be sensitive to model misspecification and censored observations; many analysts have instead used methods that exhibit strong bias. We develop an estimator with a regularization scheme to cope with stochastic delays, which we term the robust incidence deconvolution estimator. We compare the method to existing estimators in a simulation study, measuring accuracy in a variety of experimental conditions. We then use the method to study COVID-19 records in the United States, highlighting its stability in the face of misspecification and right censoring. To implement the robust incidence deconvolution estimator, we release incidental, a ready-to-use R implementation of our estimator that can aid ongoing efforts to monitor the COVID-19 pandemic.

Background: A major epidemic of COVID-19 caused by the Delta variant (B.1.617.2) occurred in India from March to July 2021, resulting in 19 million documented cases. Given the limited healthcare and testing capacities, the actual number of infections is likely to have been greater than reported, and several modelling studies and excess mortality research indicate that this epidemic involved substantial morbidity and mortality. Methods: To estimate the incidence during this epidemic, we used border entry screening data in Japan to estimate the daily incidence and cumulative incidence of COVID-19 infection in India. Analysing the results of mandatory testing among non-Japanese passengers entering Japan from India, we calculated the prevalence and then backcalculated the incidence in India from February 28 to July 3, 2021. Results: The estimated number of infections ranged from 448 to 576 million people, indicating that 31.8% (95% confidence interval (CI): 26.1, 37.7) – 40.9% (95% CI: 33.5, 48.4) of the population in India had experienced COVID-19 infection from February 28 to July 3, 2021. In addition to obtaining cumulative incidence that was consistent with published estimates, we showed that the actual incidence of COVID-19 infection during the 2021 epidemic in India was approximately 30 times greater than that based on documented cases, giving a crude infection fatality risk of 0.47%. Adjusting for test-negative certificate before departure, the quality control of which was partly questionable, the cumulative incidence can potentially be up to 2.3–2.6 times greater than abovementioned estimates. Conclusions: Our estimate of approximately 32–41% cumulative infection risk from February 28 to July 3, 2021 is roughly consistent with other published estimates, and they can potentially be greater, given an exit screening before departure. The present study results suggest the potential utility of border entry screening data to backcalculate the incidence in countries with limited surveillance capacity owing to a major surge in infections. [ABSTRACT FROM AUTHOR]

This note addresses the question of optimally estimating a linear functional of an object acquired through linear observations corrupted by random noise, where optimality pertains to a worst-case setting tied to a symmetric, convex and closed model set containing the object. It complements the article 'Statistical Estimation and Optimal Recovery' published in the Annals of Statistics in 1994. There, Donoho showed (among other things) that, for Gaussian noise, linear maps provide near-optimal estimation schemes relatively to a performance measure relevant in Statistical Estimation. Here, we advocate for a different performance measure arguably more relevant in Optimal Recovery. We show that, relatively to this new measure, linear maps still provide near-optimal estimation schemes even if the noise is merely log-concave. Our arguments, which make a connection to the deterministic noise situation and bypass properties specific to the Gaussian case, offer an alternative to parts of Donoho's proof. [ABSTRACT FROM AUTHOR]

Background: The number of people undergoing voluntary HIV testing has abruptly decreased since 2020. The geographical heterogeneity of HIV infection and the impact of COVID-19 on the diagnosis of HIV at regional level are important to understand. This study aimed to estimate the HIV incidence by geographical region and understand how the COVID-19 pandemic influenced diagnosis of HIV. Methods: We used an extended back-calculation method to reconstruct the epidemiological dynamics of HIV/AIDS by geographical region. We used eight regions: Tokyo, the capital of Japan, Hokkaido plus Tohoku, Kanto plus Koshinetsu (excluding Tokyo), Hokuriku, Tokai, Kinki, Chugoku plus Shikoku, and Kyushu plus Okinawa. Four different epidemiological measurements were evaluated: (i) estimated HIV incidence, (ii) estimated rate of diagnosis, (iii) number of undiagnosed HIV infections, and (iv) proportion of HIV infections that had been diagnosed. Results: The incidence of HIV/AIDS during the COVID-19 pandemic from 2020 to 2022 increased in all regions except Kanto/Koshinetsu (51.3 cases/year), Tokyo (183.9 cases/year), Hokuriku (1.0 cases/year), and Tokai (43.1 cases/year). The proportion of HIV infections that had been diagnosed only exceeded 90% in Tokyo (91.7%, 95% confidence interval (CI): 90.6, 93.3), Kanto/Koshinetsu (91.0%, 95% CI: 87.3, 97.8), and Kinki (92.5%, 95% CI: 90.4, 95.9). The proportion of infections that had been diagnosed was estimated at 83.3% (95% CI: 75.1, 98.7) in Chugoku/Shikoku and 80.5% (95% CI: 73.9, 91.0) in Kyusyu/Okinawa. Conclusions: In urban regions with major metropolitan cities, including Tokyo, Kinki, and Kanto/Koshinetsu, the number of undiagnosed HIV infections is substantial. However, the proportion of undiagnosed infections was estimated to be smaller than in other regions. The diagnosed proportion was the lowest in Kyusyu/Okinawa (80.5%), followed by Chugoku/Shikoku and Hokkaido/Tohoku. The level of diagnosis in those regional prefectures may have been more influenced and damaged by the COVID-19 pandemic than in urban settings.

A single-photon avalanche diode (SPAD) is a photon-counting sensor renowned for its exceptional single-photon sensitivity. One significant feature of SPADs is their non-linear response to light, making them ideal for high-dynamic range imaging applications. In SPAD imaging, the photon detection mode, which depends on the quenching method employed, is crucial for optimizing image quality and dynamic range. This paper examines the free-running and gating imaging modes, evaluating their impacts on photon capture and saturation limits. Given that the number of incident photons follows a Poisson distribution, we introduce an innovative imaging-quenching model based on statistical mathematics. We designed and fabricated two SPAD imaging sensors using 180 nm CMOS technology. Image processing and evaluation were conducted using a mapping method. Our results show that in low-light conditions, the gating mode surpasses the free-running mode in the signal-to-noise ratio (SNR). However, the free-running mode exhibits a saturation limit of more than an order of magnitude higher than that of the gating mode, demonstrating its superior capability to handle a broader range of light intensities. This paper provides a thorough analysis of the differences between the two imaging methods, incorporating the theoretical mathematical model, circuit characteristics, and computed imaging quality.

In addition to dynamic methods, purely statistical models, i.e., findings from the statistical analysis of the existing measured database, also play an important role in predicting the different characteristics of climate elements. In our article, we try to estimate the monthly amount of global radiation in each day of the month. In our previous articles, we presented the sliding-average model developed for estimating the average or amount of a climatic element, measured over a time interval, from within the interval. A version of this model for estimating the end-of-interval sums, the sliding-sum model, was used to estimate the amount of monthly global radiation. After generating the characteristics required for the estimation and analyzing their properties, we examined the errors of the performed estimation. Our model can also help solar energy users create the schedule.

A new approach to anomaly detection termed “anomaly detection with designable generative adversarial network (Ano-DGAN)” is proposed, which is a series connection of a designable generative adversarial network and anomaly detection with a generative adversarial network. The proposed Ano-DGAN, based on a deep neural network, overcomes the limitations of abnormal data collection when performing anomaly detection. In addition, it can perform statistical diagnosis by identifying the healthy range of each design variable without a massive amount of initial data. A model was constructed to simulate a high-pressure liquefied natural gas pipeline for data collection and the determination of the critical design variables. The simulation model was validated and compared with the failure mode and effect analysis of a real pipeline, which showed that stress was concentrated in the weld joints of the branch pipe. A crack-growth degradation factor was applied to the weld, and anomaly detection was performed. The performance of the proposed model was highly accurate compared with that of other anomaly detection models, such as support vector machine, 1D convolutional neural network, and long short-term memory. The results provided a statistical estimate of the design variable ranges and were validated statistically, indicating that the diagnosis was acceptable. [ABSTRACT FROM AUTHOR]

Armed conflict, displacement and food insecurity have affected Adamawa, Borno, and Yobe states of northeast Nigeria (population ≈ 12 million) since 2009. Insecurity escalated in 2013 to 2015, but the humanitarian response was delayed and the crisis’ health impact was unquantified due to incomplete death registration and limited ground access. We estimated mortality attributable to this crisis using a small-area estimation approach that circumvented these challenges. We fitted a mixed effects model to household mortality data collected as part of 70 ground surveys implemented by humanitarian actors. Model predictors, drawn from existing data, included livelihood typology, staple cereal price, vaccination geocoverage, and humanitarian actor presence. To project accurate death tolls, we reconstructed population denominators based on forced displacement. We used the model and population estimates to project mortality under observed conditions and varying assumed counterfactual conditions, had there been no crisis, with the difference providing excess mortality. Death rates were highly elevated across most ground surveys, with net negative household migration. Between April 2016 and December 2019, we projected 490,000 excess deaths (230,000 children under 5 y) in the most likely counterfactual scenario, with a range from 90,000 (best-case) to 550,000 (worst-case). Death rates were two to three times higher than counterfactual levels, double the projected national rate, and highest in 2016 to 2017. Despite limited scope (we could not study the situation before 2016 or in neighboring affected countries), our findings suggest a staggering health impact of this crisis. Further studies to document mortality in this and other crises are needed to guide decision-making and memorialize their human toll. [ABSTRACT FROM AUTHOR]

The principal goal of this article is to establish a methodological approach to computational statistical analysis for a new in-homogeneous stochastic diffusion process with mean function corresponds to the Schumacher curve. Firstly, we analyse the principal probabilistic characteristics like the functions of the mean and the transition probability density of the process. So, the parameters estimation appearing in the current process are determined by the maximum likelihood approach with discrete sampling. Finally, so as to provide the performance of the proposed process, we will apply these computational statistical results to simulated data based on a discretization of the analytical expression of the process by justifying the fit and prediction possibilities. [ABSTRACT FROM AUTHOR]

The purpose of the present study was to develop a transmission model of COVID-19 cases with and without a contact history to understand the meaning of the proportion of infected individuals with a contact history over time. We extracted epidemiological information regarding the proportion of coronavirus disease 2019 (COVID-19) cases with a contact history and analyzed incidence data stratified by the presence of a contact history in Osaka from January 15 to June 30, 2020. To clarify the relationship between transmission dynamics and cases with a contact history, we used a bivariate renewal process model to describe transmission among cases with and without a contact history. We quantified the next-generation matrix as a function of time; thus, the instantaneous (effective) reproduction number was calculated for different periods of the epidemic wave. We objectively interpreted the estimated next-generation matrix and replicated the proportion of cases with a contact p(t) over time, and we examined the relevance to the reproduction number. We found that p(t) does not take either the maximum or minimum value at a threshold level of transmission with R(t)=1.0. With R(t) < 1 (subcritical level), p(t) was a decreasing function of R(t). Qualitatively, the minimum p(t) was seen in the domain with R(t) > 1. An important future implication for use of the proposed model is to monitor the success of ongoing contact tracing practice. A decreasing signal of p(t) reflects the increasing difficulty of contact tracing. The present study findings indicate that monitoring p(t) would be a useful addition to surveillance.

Abstract Background It has been descriptively argued that the case fatality risk (CFR) of coronavirus disease (COVID-19) is elevated when medical services are overwhelmed. The relationship between CFR and pressure on health-care services should thus be epidemiologically explored to account for potential epidemiological biases. The purpose of the present study was to estimate the age-dependent CFR in Tokyo and Osaka over time, investigating the impact of caseload demand on the risk of death. Methods We estimated the time-dependent CFR, accounting for time delay from diagnosis to death. To this end, we first determined the time distribution from diagnosis to death, allowing variations in the delay over time. We then assessed the age-dependent CFR in Tokyo and Osaka. In Osaka, the risk of intensive care unit (ICU) admission was also estimated. Results The CFR was highest among individuals aged 80 years and older and during the first epidemic wave from February to June 2020, estimated as 25.4% (95% confidence interval [CI] 21.1 to 29.6) and 27.9% (95% CI 20.6 to 36.1) in Tokyo and Osaka, respectively. During the fourth wave of infection (caused by the Alpha variant) in Osaka the CFR among the 70s and ≥ 80s age groups was, respectively, 2.3 and 1.5 times greater than in Tokyo. Conversely, despite the surge in hospitalizations, the risk of ICU admission among those aged 80 and older in Osaka decreased. Such time-dependent variation in the CFR was not seen among younger patients

Pesticide surface water monitoring data have rarely been used as the only quantitative measure of exposure because the available monitoring data for most pesticides has not been considered robust enough for direct use in pesticide exposure assessments due to infrequent sampling. The cost of daily sample collection and analysis prohibits frequent sampling for most monitoring programs. In this context, a common question raised in assessments is how likely peak concentrations (i.e., annual maxima) may be missed if sampling intervals are longer than daily. The US Geological Survey developed the statistical model "seasonal wave with streamflow adjustment and extended capability" (SEAWAVE–QEX) to address the need to estimate infrequently occurring pesticide concentrations, such as annual maximum daily concentrations, for sites with nondaily monitoring data. This study compares the results of two postprocessing methods and evaluates the capability of SEAWAVE–QEX to estimate annual maximum concentrations of three commonly used herbicides and one metabolite in a catchment in Belgium. The study concludes that the appropriateness of using SEAWAVE–QEX to estimate annual maximum concentrations depends on pesticide characteristics and use and that the model can be particularly sensitive to nonflow correlated exposure events (e.g., point source contributions or drift). Integr Environ Assess Manag 2023;19:513–526. © 2022 Stone Environmental and Bayer AG Crop Science Division. Integrated Environmental Assessment and Management published by Wiley Periodicals LLC on behalf of Society of Environmental Toxicology & Chemistry (SETAC). Key Points: SEAWAVE–QEX performance was variable and dependent on pesticide characteristics and use; however, a relationship between pesticide characteristics and model performance could not be identified, with the best performance observed for the subsurface transport driven metabolite.Choice of a statistical postprocessing approach of SEAWAVE–QEX results has a significant impact on the model's estimated 1‐day annual maximum herbicide concentrations.Removing independently identified point‐source‐driven high concentration events from the input data improved SEAWAVE–QEX 1‐day annual maximum concentration estimates.The SEAWAVE–QEX statistical model was used to estimate 1‐day annual maximum concentrations for three herbicides and a metabolite in a stream draining a small agricultural catchment. [ABSTRACT FROM AUTHOR]

In addition to dynamic methods, purely statistical models, i.e., findings from the statistical analysis of the existing measured database, also play an important role in predicting the different characteristics of climate elements. In our article, we try to estimate the monthly amount of global radiation in each day of the month. In our previous articles, we presented the sliding-average model developed for estimating the average or amount of a climatic element, measured over a time interval, from within the interval. A version of this model for estimating the end-of-interval sums, the sliding-sum model, was used to estimate the amount of monthly global radiation. After generating the characteristics required for the estimation and analyzing their properties, we examined the errors of the performed estimation. Our model can also help solar energy users create the schedule. [ABSTRACT FROM AUTHOR]

The judges for the 2022 Daniel H. Wagner Prize for Excellence in the Practice of Advanced Analytics and Operations Research selected the four finalist papers featured in this special issue of the INFORMS Journal on Applied Analytics (IJAA). The prestigious Wagner Prize--awarded for achievement in implemented operations research, management science, and advanced analytics--emphasizes the quality and originality of mathematical models along with clarity of written and oral exposition. This year's winning application describes the design and deployment of a generalized synthetic control, a powerful and innovative statistical method for identifying, in a noisy environment, retailing innovations that produce a small percentage improvement in a large volume of sales for Anheuser Busch Inbev. The remaining three papers describe an inverse control approach to allocating lung transplants that best meets targeted outcomes and has been implemented as the national lung allocation policy on March 9, 2023, across the United States; a human-centric, optimized parcel delivery system developed for Deutsche Post that saves money while meeting constraints learned dynamically from driver behavior; and an AI-based system developed for Alibaba that learns supplier behavior to improve replenishment ordering and inventory control. [ABSTRACT FROM AUTHOR]

The article gives information about the attempt made to select configurations, train and test artificial neural networks for predicting yields of grain crops considering of climate changes. Peculiarities of agricultural production require constant improvement of methods for analyzing crop yields, time series, and longterm climatic characteristics. Preliminary statistical evaluation of the considered time series made it possible to identify certain patterns. Time series were divided into four intervals: for building a network, its training, testing and control. During the construction of artificial neural networks, three models were used: MLP - multilayer perceptron, RBF - r adial basis functions and GRNN - g eneralized regression neural network. Based on the results of the construction, the best model was chosen. The sum of active air temperatures and the sum of precipitation for the growing season was used for artificial neural networks at the input, and the crop yield was used at the output. The use of sets of neural systems, generated automatically, contributed to the effective forecasting of crop yields based on the analysis of climate data. As a result, according to the selected model, a yield forecast was made for the coming years considering climatic characteristics.

This work presents a Bayesian statistical approach to the saliency map estimation problem. More specifically, we formalize the saliency map estimation issue in the fully automatic Markovian framework. The major and original contribution of the proposed Bayesian–Markov model resides in the exploitation of a pixel pairwise modeling and a likelihood model based on a parametric mixture of two different class-conditional likelihood distributions whose parameters are adaptively and previously estimated for each image. This allows us to adapt our saliency estimation model to the specific characteristics of each image of the dataset and to provide a nearly parameter-free—hence dataset-independent—unsupervised saliency map estimation procedure. In our case, the parameters of the likelihood model are all estimated under the principles of the iterative conditional estimation framework. Once the estimation step is completed, the MPM (maximum posterior marginal) solution of the saliency map (which we show as particularly suitable for this type of estimation), is then estimated by a stochastic sampling scheme approximating the posterior distribution (whose parameters were previously estimated). This unsupervised data-driven Markovian framework overcomes the limitations of current ad hoc or supervised energy-based or Markovian models that often involve many parameters to adapt and that are finely tuned for each different benchmark database. Experimental results show that the proposed algorithm performs favorably against state-of-the-art methods and turns out to be particularly stable across a wide variety of benchmark datasets. [ABSTRACT FROM AUTHOR]

Generative artificial intelligence has made significant strides, producing text indistinguishable from human prose and remarkably photorealistic images. Automatically measuring how close the generated data distribution is to the target distribution is central to diagnosing existing models and developing better ones. We present MAUVE, a family of comparison measures between pairs of distributions such as those encountered in the generative modeling of text or images. These scores are statistical summaries of divergence frontiers capturing two types of errors in generative modeling. We explore three approaches to statistically estimate these scores: vector quantization, non-parametric estimation, and classifier-based estimation. We provide statistical bounds for the vector quantization approach. Empirically, we find that the proposed scores paired with a range of f-divergences and statistical estimation methods can quantify the gaps between the distributions of humanwritten text and those of modern neural language models by correlating with human judgments and identifying known properties of the generated texts. We demonstrate in the vision domain that MAUVE can identify known properties of generated images on par with or better than existing metrics. In conclusion, we present practical recommendations for using MAUVE effectively with language and image modalities. [ABSTRACT FROM AUTHOR]

Whilst several methods exist to provide sample estimates of the probability distribution function at several points, for the probability density of continuous stochastic variables, only a gross representation through the histogram is typically used. It is shown that the newly introduced concept of knowable moments (K-moments) can provide smooth empirical representations of the distribution function, which in turn can yield point and interval estimates of the density function at a large number of points or even at any arbitrary point within the range of the available observations. The proposed framework is simple to apply and is illustrated with several applications for a variety of distribution functions. [ABSTRACT FROM AUTHOR]

Applying a statistical test to check statistical assumptions offers a positive or negative response regarding the veracity of the issued hypothesis. In case of variance analysis it’s necessary to apply a post hoc test to determine differences within the group. Statistical estimation using confidence levels provides more information than a statistical test, it shows the high degree of uncertainty resulting from small samples and builds conclusions in terms of "marginally significant" or "almost significant (p being close to 0,05) . The case study shows how the statistical estimation completes the application form the analysis of variance test and Tukey test.

Background: A measles outbreak involving 60 cases occurred in Yamagata, Japan in 2017. Using two different mathematical models for different datasets, we aimed to estimate measles transmissibility over time and explore any heterogeneous transmission patterns. Methods: The first model relied on the temporal distribution for date of illness onset for cases, and a generation-dependent model was applied to the data. Another model focused on the transmission network. Using the illness-onset date along with the serial interval and geographical location of exposure, we reconstructed a transmission network with 19 unknown links. We then compared the number of secondary transmissions with and without clinical symptoms or laboratory findings. Results: Using a generation-dependent model (assuming three generations other than the index case), the reproduction number (R) over generations 0, 1, and 2 were 25.3, 1.3, and

This paper studies the consequences of imperfect information for the precision of stochastic optimization. In particular, it is assumed that the stochastic characteristics of an optimization problem depend on unknown parameters estimated from available data. First, a theoretical result is presented, showing that consistent parameters estimation leads to consistent optimization. Further, a type of the studied models is specified; it is assumed that the random variables present in the optimization problem are influenced by covariates. This influence is expressed via a parametric regression model, whose parameters have to be estimated and used instead of the unknown correct parameters values. The objective is then to explore, with the aid of simulations, the imprecision of the optimization based on these estimates. Several types of regression models are recalled, the variability of estimates and the related precision of sub-optimal solutions is studied in detail on an example dealing with optimal maintenance. The impact of random right-censoring on the deterioration of precision is studied as well.

Coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), has rapidly spread across the globe. The variant of concern (VOC) 202012/01 (B.1.1.7, also known as the alpha variant) bearing the N501Y mutation emerged in late 2020. VOC 202012/01 was more transmissible than existing SARS-CoV-2 variants and swiftly became dominant in many regions. More than 150 cases of VOC 202012/01 were reported in Japan by 26 February 2021. During the very early stage of introduction, only a subset arose from domestic transmission. If the reproduction number R (i.e., the average number of secondary transmission events caused by a single primary case) is greater than 1, the corresponding proportion should converge to 1 in a short period of time, and thus it is critical to understand the transmissibility of VOC 202012/01 based on travel history information. The present study aimed to estimate R of VOC 202012/01 using overseas travel history information. A mathematical model was developed to capture the relationship between travel history and R. We obtained travel history data for each confirmed case of VOC 202012/01 infection from 26 December 2020 to 26 February 2021. Maximum likelihood estimation was used to estimate R, accounting for right censoring during real-time estimation. In the baseline scenario, R was estimated at 2.11 (95% confidence interval: 1.63, 2.94). By 26 February 2021, an average of nine generations had elapsed since the first imported case. If the generation time of VOC 202012/01 was assumed to be longer, R was increased, consistent with estimates of R from case data. The estimated R of VOC 202012/01 in Japan exceeded 1 on 26 February 2021, suggesting that domestic transmission events caused a major epidemic. Moreover, because our estimate of R was dependent on generation time and ascertainment biases, continuous monitoring of contact tracing data is crucial to decipher the mechanisms of increased VOC 202012/01 transmissibility.

Model a network as an edge-weighted graph, where the (unknown) weight we of edge e indicates the frequency of observed interactions, and over time t we observe a Poisson (twe) number of interactions across edges e. How should we estimate some given statistic of the underlying network? This leads to wide-ranging and challenging problems, on which this article makes only partial progress.

Variational representations of divergences and distances between high-dimensional probability distributions offer significant theoretical insights and practical advantages in numerous research areas. Recently, they have gained popularity in machine learning as a tractable and scalable approach for training probabilistic models and for statistically differentiating between data distributions. Their advantages include: 1) They can be estimated from data as statistical averages. 2) Such representations can leverage the ability of neural networks to efficiently approximate optimal solutions in function spaces. However, a systematic and practical approach to improving the tightness of such variational formulas, and accordingly accelerate statistical learning and estimation from data, is currently lacking. Here we develop such a methodology for building new, tighter variational representations of divergences. Our approach relies on improved objective functionals constructed via an auxiliary optimization problem. Furthermore, the calculation of the functional Hessian of objective functionals unveils the local curvature differences around the common optimal variational solution; this quantifies and orders the tightness gains between different variational representations. Finally, numerical simulations utilizing neural network optimization demonstrate that tighter representations can result in significantly faster learning and more accurate estimation of divergences in both synthetic and real datasets (of more than 1000 dimensions), often accelerated by nearly an order of magnitude. [ABSTRACT FROM AUTHOR]

This study attempts to explore the influence of observations in a time series or a discrete time signal. The goal is to detect abnormal observations from a frequency domain point of view, while the most of relevant studies have been done from a time domain point of view. The concept of the influence function in the field of robust statistics is borrowed to identify influential observations in a time series. An empirical version of the influence function on the discrete Fourier transform of a time series is designed and subsequently a statistic is proposed to identify influential observations of a time series from the frequency domain point of view. Though the proposed statistic is simple enough to be calculated with simple arithmetic operations, case studies show that the proposed method is capable of identifying influential or abnormal observations of a time series. By identifying influential or abnormal observations, we would be able to gain a better understanding of the nature of a time series and to control possible future influential observations. [ABSTRACT FROM AUTHOR]

Randomly computed elliptic curves over prime fields are desirable for implementation in strategic or military grade cryptosystems. A randomly computed elliptic curve eliminates the possibility of exhibiting any pre‐studied value or special structure of the curve parameters which may be doubted to have intentionally non‐disclosed vulnerability. However, computing elliptic curves randomly over large prime fields for cryptography is an intricate task which requires reasonably high computational resources and time. These computational resources are considered in terms of the number of CPU clock cycles and the number of attempts or searches made in the security parameter space of elliptic curve. The estimate of the number of CPU clock cycles helps in determining processor requirements whereas the number of attempts or searches help to decide the number of CPU cores for speeding up the curve generation process. We present efficient statistical estimates of computational resources to randomly compute cryptographically safe elliptic curve over desired prime field size using a standard procedure. [ABSTRACT FROM AUTHOR]

El conocimiento sobre lo saludable de los productos orgánicos abre posibilidades para el mejoramiento de la alimentación y posible reducción de enfermedades de tipo mórbidas. Este manuscrito analiza el efecto halo producido en la percepción del café orgánico en relación con cinco aspectos: sabor, nutrición, percepción saludable, natural y residual. Se aplican cinco diseños de experimentos factoriales, considerando entre otros, el factor producto/etiqueta de café cuyos niveles son: A, B, orgánico, regular y C; sin informar que A es café regular y B es café orgánico y C es un café regular. Los productos etiquetados como orgánicos se perciben más saludables, nutritivos y de mejores sabores; además, se encuentra que la etiqueta en el empaque del café orgánico genera un efecto halo al predisponer su percepción con mejor sabor y percepción saludable sobre los demás, aunque uno de ellos se enmascara con otra etiqueta siendo igual. La producción orgánica de café ha generado valor agregado, incrementando el mercado saludable y ambientalista, aprovechando la identidad alcanzada por los productos orgánicos, como se demuestra en la investigación por su alto valor agregado percibido, esto incrementa las posibilidades del marketing, ventas y nuevos desarrollos. The knowledge of the healthiness of organic products opens possibilities for the improvement of the diet and possible reduction of morbid diseases. The purpose of this study is to analyze the effect on the perceived value of coffee labeled organic with respect to aspects such as taste and nutritional perception. Five factorial experiment designs are applied, considering among others, the factor product/coffee label, whose levels are: A, B, organic, regular and C; without informing that A is regular coffee and B is organic coffee and C is regular coffee. The healthiest, most nutritious, and best tasting organic products are perceived; in addition, it is found that the label on the packaging of organic coffee generates a halo effect by predisposing its perception with better flavor and healthy perception over the other types of labeled products, even though one of them is masked with another label being from the same organic. The organic production of coffee has generated added value, increasing the healthy and environmental market, taking advantage of the identity achieved by organic products, as demonstrated in the research by its high perceived added value, contributing in the increase of possibilities of marketing, sales and new developments.

This paper presents a modeling framework for joint estimation of a host vehicle state and a map of the road based on global navigation satellite system (GNSS) and camera measurements. We model the road using a spline representation based on lower-dimensional Bézier curves parametrized in generalized endpoints (GEPs) with implicit guarantees of continuous lane boundaries. We model the GEPs by a parameter vector having a Gaussian prior representing the uncertainty of the prior map, and provide a systematic way of defining this prior from generic map representations. Both GNSS and camera measurements, such as lane-mark measurements, have noise characteristics that vary in time. To adapt to the changing noise levels and hence improve positioning performance, we formulate the problem as a joint vehicle state, map parameter, and noise covariance estimation problem and present two noise-adaptive linear-regression Kalman filters (LRKFs); (i) an interacting multiple-model (IMM) LRKF and (ii) a variational-Bayes (VB) LRKF. We conduct a Monte-Carlo study and compare the two approaches in terms of estimation precision and computation times. Embedded implementations in an automotive-grade dSpace Micro Autobox-II indicate the real-time validity of both approaches, with turn-around times of between 2– 80 ms , depending on the problem size and if the map is updated. The results indicate that while the IMM-LRKF shows marginally better estimation accuracy, the VB-LRKF is at least a factor of 2 faster. [ABSTRACT FROM AUTHOR]

The paper proposes a new maintenance model for a repairable system. The system is inspected at deterministic times. If a failure occurs between two successive inspections, a minimal repair is performed. At inspection times, two types of maintenance actions can be performed. The first originality of the paper is to consider that the choice between these two types of maintenance depends on the fact that there has been a failure or not since the previous inspection. If no failure has occurred since the last inspection, the maintenance, simply denoted PM, corresponds to a basic preventive maintenance action, whose effect is planned by the initial maintenance plan. If at least one failure has occurred since the last inspection, the maintenance, denoted PCM, includes both the preventive action planned by the initial maintenance plan, and a specific corrective action which aims to fix the defects that have happened since the last inspection. Both types of maintenances are assumed to be imperfect, of the ARA (Arithmetic Reduction of Age) type, with different parameters. Moreover, the paper derives several statistical inference methods for this model. The usual case of complete data is first considered. The second originality of the paper is to analyse interval-censored data in this imperfect maintenance framework. This situation holds when failure times are not observed, and the only information is the fact that failures have occurred or not between two inspections. The quality of the proposed methods is assessed on simulated data. Finally, a real dataset from a gas transmission system operator is provided and analysed with the proposed approach. • A model for repairable systems with corrective maintenance and two types of imperfect preventive maintenance is proposed. • Maintenance types depend on the fact that there has been a failure or not since the previous inspection. • Statistical inference methods are developed for complete and interval-censored data. • A real dataset from a gas transmission system operator is provided and analysed with the proposed approach. [ABSTRACT FROM AUTHOR]