Timely Reporting of Heavy Hitters Using External Memory

Given an input stream S of size N , a ɸ-heavy hitter is an item that occurs at least ɸN times in S . The problem of finding heavy-hitters is extensively studied in the database literature. We study a real-time heavy-hitters variant in which an element must be reported shortly after we see its T = ɸ N-th occurrence (and hence it becomes a heavy hitter). We call this the Timely Event Detection ( TED ) Problem. The TED problem models the needs of many real-world monitoring systems, which demand accurate (i.e., no false negatives) and timely reporting of all events from large, high-speed streams with a low reporting threshold (high sensitivity). Like the classic heavy-hitters problem, solving the TED problem without false-positives requires large space (Ω (N) words). Thus in-RAM heavy-hitters algorithms typically sacrifice accuracy (i.e., allow false positives), sensitivity, or timeliness (i.e., use multiple passes). We show how to adapt heavy-hitters algorithms to external memory to solve the TED problem on large high-speed streams while guaranteeing accuracy, sensitivity, and timeliness. Our data structures are limited only by I/O-bandwidth (not latency) and support a tunable tradeoff between reporting delay and I/O overhead. With a small bounded reporting delay, our algorithms incur only a logarithmic I/O overhead. We implement and validate our data structures empirically using the Firehose streaming benchmark. Multi-threaded versions of our structures can scale to process 11M observations per second before becoming CPU bound. In comparison, a naive adaptation of the standard heavy-hitters algorithm to external memory would be limited by the storage device’s random I/O throughput, i.e., ≈100K observations per second.

Data Mining in Automotive Warranty Analysis

<p>This thesis is about data mining in automotive warranty analysis, with an emphasis on modeling the mean cumulative warranty cost or number of claims (per vehicle). In our study, we deal with a type of truncation that is typical for automotive warranty data, where the warranty coverage and the resulting warranty data are limited by age and mileage. Age, as a function of time, is known for all sold vehicles at all time. However, mileage is only observed for a vehicle with at least one claim and only at the time of the claim. To deal with this problem of incomplete mileage information, we consider a linear approach and a piece-wise linear approach within a nonparametric framework. We explore the univariate case, as well as the bivariate case. For the univariate case, we evaluate the mean cumulative warranty cost and its standard error as a function of age, a function of mileage, and a function of actual (calendar) time. For the bivariate case, we evaluate the mean cumulative warranty cost as a function of age and mileage. The effect of reporting delay of claim and several methods for making prediction are also considered. Throughout this thesis, we illustrate the ideas using examples based on real data.</p>

Data Mining in Automotive Warranty Analysis

<p>This thesis is about data mining in automotive warranty analysis, with an emphasis on modeling the mean cumulative warranty cost or number of claims (per vehicle). In our study, we deal with a type of truncation that is typical for automotive warranty data, where the warranty coverage and the resulting warranty data are limited by age and mileage. Age, as a function of time, is known for all sold vehicles at all time. However, mileage is only observed for a vehicle with at least one claim and only at the time of the claim. To deal with this problem of incomplete mileage information, we consider a linear approach and a piece-wise linear approach within a nonparametric framework. We explore the univariate case, as well as the bivariate case. For the univariate case, we evaluate the mean cumulative warranty cost and its standard error as a function of age, a function of mileage, and a function of actual (calendar) time. For the bivariate case, we evaluate the mean cumulative warranty cost as a function of age and mileage. The effect of reporting delay of claim and several methods for making prediction are also considered. Throughout this thesis, we illustrate the ideas using examples based on real data.</p>

Monday effect on confirmed cases of COVID-19 in Japan

We examined the phenomenon of fewer new confirmed cases on Monday in Japan, which we refer to as the Monday effect. In Japan, prefectures aggregate and announce the number of daily confirmed cases. We analyzed the impact of this effect in each prefecture. The effect is mainly found in prefectures with populations of 2 million or more. This effect is also constantly observed in the three major metropolitan areas in Japan. However, the magnitude of the observed effect is uncorrelated with both the number of positives per 1,000 people and the population size. Our results suggest that the reporting delay occurs in prefectures above a specific size, but the magnitude of the delay differs among prefectures. We consider two possible explanations for this effect: 1) delays caused by the administrative system. 2) fewer tests are conducted on the previous day. Our results indicate that delays are caused by the administrative system in some prefectures and that some prefectures with larger populations are less likely to conduct screenings on holidays.

Bayesian back-calculation and nowcasting for line list data during the COVID-19 pandemic

Surveillance is critical to mounting an appropriate and effective response to pandemics. However, aggregated case report data suffers from reporting delays and can lead to misleading inferences. Different from aggregated case report data, line list data is a table contains individual features such as dates of symptom onset and reporting for each reported case and a good source for modeling delays. Current methods for modeling reporting delays are not particularly appropriate for line list data, which typically has missing symptom onset dates that are non-ignorable for modeling reporting delays. In this paper, we develop a Bayesian approach that dynamically integrates imputation and estimation for line list data. Specifically, this Bayesian approach can accurately estimate the epidemic curve and instantaneous reproduction numbers, even with most symptom onset dates missing. The Bayesian approach is also robust to deviations from model assumptions, such as changes in the reporting delay distribution or incorrect specification of the maximum reporting delay. We apply the Bayesian approach to COVID-19 line list data in Massachusetts and find the reproduction number estimates correspond more closely to the control measures than the estimates based on the reported curve.

Delayed emergency healthcare seeking behaviour by Dutch emergency department visitors during the first COVID-19 wave: a mixed methods retrospective observational study

Background Emergency department (ED) visits due to non-coronavirus disease 2019 (COVID-19) conditions have drastically decreased since the outbreak of the COVID-19 pandemic. This study aimed to identify the magnitude, characteristics and underlying motivations of ED visitors with delayed healthcare seeking behaviour during the first wave of the pandemic. Methods Between March 9 and July 92,020, adults visiting the ED of an academic hospital in the East of the Netherlands received an online questionnaire to collect self-reported data on delay in seeking emergency care and subsequent motivations for this delay. Telephone interviews were held with a subsample of respondents to better understand the motivations for delay as described in the questionnaire. Quantitative data were analysed using descriptive statistics. Qualitative data were thematically analysed. Results One thousand three hundred thirty-eight questionnaires were returned (34.0% response). One in five respondents reported a delay in seeking emergency care. Almost half of these respondents (n = 126; 45.4%) reported that the pandemic influenced the delay. Respondents reporting delay were mainly older adults (mean 61.6; ±13.1 years), referred to the ED by the general practitioner (GP; 35.1%) or a medical specialist (34.7%), visiting the ED with cardiac problems (39.7%). The estimated median time of delay in receiving ED care was 3 days (inter quartile range 8 days). Respectively 46 (16.5%) and 26 (9.4%) respondents reported that their complaints would be either less severe or preventable if they had sought for emergency care earlier. Delayed care seeking behaviour was frequently motivated by: fear of contamination, not wanting to burden professionals, perceiving own complaints less urgent relative to COVID-19 patients, limited access to services, and by stay home instructions from referring professionals. Conclusions A relatively large proportion of ED visitors reported delay in seeking emergency care during the first wave. Delay was often driven by misperceptions of the accessibility of services and the legitimacy for seeking emergency care. Public messaging and close collaboration between the ED and referring professionals could help reduce delayed care for acute needs during future COVID-19 infection waves.

OPINI AUDIT, AUDIT REPORT LAG, REPORTING DELAY, KAP, DAN EPS SEBAGAI DETERMINAN HARGA SAHAM

The information about company’s financial performance can be obtained by investor in company’s financial report. Some of the informations are audit opinion, audit report lag, reporting delay, KAP, and EPS. That informations are used by investor to do an investment analysis through financial report that can influence investment decision. Investor’s investment decision can influence stock price change. The purposes of this research are to determine the simultan and partial effect of audit opinion, audit report lag, reporting delay, KAP, and EPS on stock price. Research type that is used quantitative method and descriptive approach. Research data are financial reports and stock prices with banking companies listed on Indonesia Stock Exchange (BEI) within year of 2016-2018 as a population. Purposive sampling is used as a sampling method so obtained 23 samples. The research method is using multiple linier regression analysis. The results of this research are simultaneously audit opinion, audit report lag, reporting delay, KAP, and EPS have significant influence to stock prices. Partially, reporting delay and EPS have significant positive influence to stock prices. Meanwhile, audit opinion, audit report lag, and KAP have no effect on stock prices.

Analysis of the early COVID-19 epidemic curve in Germany by regression models with change points

We analysed the coronavirus disease 2019 epidemic curve from March to the end of April 2020 in Germany. We use statistical models to estimate the number of cases with disease onset on a given day and use back-projection techniques to obtain the number of new infections per day. The respective time series are analysed by a trend regression model with change points. The change points are estimated directly from the data. We carry out the analysis for the whole of Germany and the federal state of Bavaria, where we have more detailed data. Both analyses show a major change between 9 and 13 March for the time series of infections: from a strong increase to a decrease. Another change was found between 25 March and 29 March, where the decline intensified. Furthermore, we perform an analysis stratified by age. A main result is a delayed course of the pandemic for the age group 80 + resulting in a turning point at the end of March. Our results differ from those by other authors as we take into account the reporting delay, which turned out to be time dependent and therefore changes the structure of the epidemic curve compared to the curve of newly reported cases.

Delay in death reporting affects timely monitoring and modeling of the COVID-19 pandemic

This study describes the COVID-19 death reporting delay in the city of São Luís, Maranhão State, Brazil, and shows its impact on timely monitoring and modeling of the COVID-19 pandemic, while seeking to ascertain how nowcasting can improve death reporting delay. We analyzed COVID-19 death data reported daily in the Epidemiological Bulletin of the State Health Secretariat of Maranhão and calculated the reporting delay from March 23 to August 29, 2020. A semi-mechanistic Bayesian hierarchical model was fitted to illustrate the impact of death reporting delay and test the effectiveness of a Bayesian Nowcasting in improving data quality. Only 17.8% of deaths were reported without delay or the day after, while 40.5% were reported more than 30 days late. Following an initial underestimation due to reporting delay, 644 deaths were reported from June 7 to August 29, although only 116 deaths occurred during this period. Using the Bayesian nowcasting technique partially improved the quality of mortality data during the peak of the pandemic, providing estimates that better matched the observed scenario in the city, becoming unusable nearly two months after the peak. As delay in death reporting can directly interfere with assertive and timely decision-making regarding the COVID-19 pandemic, the Brazilian epidemiological surveillance system must be urgently revised and notifying the date of death must be mandatory. Nowcasting has proven somewhat effective in improving the quality of mortality data, but only at the peak of the pandemic.

Analysis of the early Covid-19 epidemic curve in Germany by regression models with change points

We analyze the Covid-19 epidemic curve from March to end of April 2020 in Germany. We use statistical models to estimate the number of cases with disease onset on a given day and use back-projection techniques to obtain the number of new infections per day. The respective time series are analyzed by a Poisson trend regression model with change points. The change points are estimated directly from the data without further assumptions. We carry out the analysis for the whole of Germany and the federal state of Bavaria, where we have more detailed data. Both analyses show a major change between March 9th and 13th for the time series of infections: from a strong increase to a stagnation or a slight decrease. Another change was found between March 24th and March 31st, where the decline intensified. These two major changes can be related to different governmental measures. On March, 11th, Chancellor Merkel appealed for social distancing in a press conference with the Robert Koch Institute (RKI) and a ban on major events with more than 1000 visitors (March 10th) was issued. The other change point at the end of March could be related to the shutdown in Germany.Our results differ from those by other authors as we take into account the reporting delay, which turned out to be time dependent and therefore changes the structure of the epidemic curve compared to the curve of newly reported cases.