arXiv - STAT - Applications最新文献

{"title":"Multiway Alignment of Political Attitudes","authors":"Letizia Iannucci, Ali Faqeeh, Ali Salloum, Ted Hsuan Yun Chen, Mikko Kivelä","doi":"arxiv-2408.00139","DOIUrl":"https://doi.org/arxiv-2408.00139","url":null,"abstract":"The related concepts of partisan belief systems, issue alignment, and\u0000partisan sorting are central to our understanding of politics. These phenomena\u0000have been studied using measures of alignment between pairs of topics, or how\u0000much individuals' attitudes toward a topic reveal about their attitudes toward\u0000another topic. We introduce a higher-order measure that extends the assessment\u0000of alignment beyond pairs of topics by quantifying the amount of information\u0000individuals' opinions on one topic reveal about a set of topics simultaneously.\u0000Our multiway alignment measure indicates how much individuals' opinions on\u0000multiple topics align into a single ideological divide. Applying this approach\u0000to legislative voting behavior reveals that parliamentary systems typically\u0000exhibit similar multiway alignment characteristics, but can change in response\u0000to shifting intergroup dynamics. In American National Election Studies surveys,\u0000our approach reveals a growing significance of party identification together\u0000with a consistent rise in multiway alignment over time. Similarly, the growing\u0000multiway alignment among topical issues in Finnish online discussions suggests\u0000a trend towards a more ideologically driven political landscape. Our case\u0000studies demonstrate that the multiway alignment measure is a versatile tool for\u0000understanding societal polarization and partisan belief systems across diverse\u0000domains.","PeriodicalId":501172,"journal":{"name":"arXiv - STAT - Applications","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141882625","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Large-scale Epidemiological modeling: Scanning for Mosquito-Borne Diseases Spatio-temporal Patterns in Brazil","authors":"Eduardo C. Araujo, Claudia T. Codeço, Sandro Loch, Luã B. Vacaro, Laís P. Freitas, Raquel M. Lana, Leonardo S. Bastos, Iasmim F. de Almeida, Fernanda Valente, Luiz M. Carvalho, Flávio C. Coelho","doi":"arxiv-2407.21286","DOIUrl":"https://doi.org/arxiv-2407.21286","url":null,"abstract":"The influence of climate on mosquito-borne diseases like dengue and\u0000chikungunya is well-established, but comprehensively tracking long-term spatial\u0000and temporal trends across large areas has been hindered by fragmented data and\u0000limited analysis tools. This study presents an unprecedented analysis, in terms\u0000of breadth, estimating the SIR transmission parameters from incidence data in\u0000all 5,570 municipalities in Brazil over 14 years (2010-2023) for both dengue\u0000and chikungunya. We describe the Episcanner computational pipeline, developed\u0000to estimate these parameters, producing a reusable dataset describing all\u0000dengue and chikungunya epidemics that have taken place in this period, in\u0000Brazil. The analysis reveals new insights into the climate-epidemic nexus: We\u0000identify distinct geographical and temporal patterns of arbovirus disease\u0000incidence across Brazil, highlighting how climatic factors like temperature and\u0000precipitation influence the timing and intensity of dengue and chikungunya\u0000epidemics. The innovative Episcanner tool empowers researchers and public\u0000health officials to explore these patterns in detail, facilitating targeted\u0000interventions and risk assessments. This research offers a new perspective on\u0000the long-term dynamics of climate-driven mosquito-borne diseases and their\u0000geographical specificities linked to the effects of global temperature\u0000fluctuations such as those captured by the ENSO index.","PeriodicalId":501172,"journal":{"name":"arXiv - STAT - Applications","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141866619","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Powerful A/B-Testing Metrics and Where to Find Them","authors":"Olivier Jeunen, Shubham Baweja, Neeti Pokharna, Aleksei Ustimenko","doi":"arxiv-2407.20665","DOIUrl":"https://doi.org/arxiv-2407.20665","url":null,"abstract":"Online controlled experiments, colloquially known as A/B-tests, are the bread\u0000and butter of real-world recommender system evaluation. Typically, end-users\u0000are randomly assigned some system variant, and a plethora of metrics are then\u0000tracked, collected, and aggregated throughout the experiment. A North Star\u0000metric (e.g. long-term growth or revenue) is used to assess which system\u0000variant should be deemed superior. As a result, most collected metrics are\u0000supporting in nature, and serve to either (i) provide an understanding of how\u0000the experiment impacts user experience, or (ii) allow for confident\u0000decision-making when the North Star metric moves insignificantly (i.e. a false\u0000negative or type-II error). The latter is not straightforward: suppose a\u0000treatment variant leads to fewer but longer sessions, with more views but fewer\u0000engagements; should this be considered a positive or negative outcome? The question then becomes: how do we assess a supporting metric's utility\u0000when it comes to decision-making using A/B-testing? Online platforms typically\u0000run dozens of experiments at any given time. This provides a wealth of\u0000information about interventions and treatment effects that can be used to\u0000evaluate metrics' utility for online evaluation. We propose to collect this\u0000information and leverage it to quantify type-I, type-II, and type-III errors\u0000for the metrics of interest, alongside a distribution of measurements of their\u0000statistical power (e.g. $z$-scores and $p$-values). We present results and\u0000insights from building this pipeline at scale for two large-scale short-video\u0000platforms: ShareChat and Moj; leveraging hundreds of past experiments to find\u0000online metrics with high statistical power.","PeriodicalId":501172,"journal":{"name":"arXiv - STAT - Applications","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-07-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141866624","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Parameters Evolution in Source-Sink Space Population Evolutionary Models","authors":"Erin Ashley, Carla Simon Sanz, Simone Servadio, Giovanni Lavezzi","doi":"arxiv-2407.21000","DOIUrl":"https://doi.org/arxiv-2407.21000","url":null,"abstract":"MOCAT-SSEM is a Source-Sink model that predicts the Low Earth Orbit (LEO)\u0000space population divided into families using a predefined set of interaction\u0000parameters. Thanks to data from the Monte Carlo version of the model\u0000(MOCAT-MC), which propagates singularly every object, it is possible to\u0000estimate such parameters, assumed as additional stochastic variables. Thus,\u0000this paper proposed a new set of parameters so that the new Source-Sink model\u0000prediction better fits the computationally expensive and accurate MOCAT-MC\u0000simulation. Estimation is performed by extracting stochastic quantities from\u0000the space population, which has been analyzed to fit common probability density\u0000functions.","PeriodicalId":501172,"journal":{"name":"arXiv - STAT - Applications","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-07-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141866621","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"AI-Powered Energy algorithmic Trading: Integrating Hidden Markov Models with Neural Networks","authors":"Tiago Monteiro","doi":"arxiv-2407.19858","DOIUrl":"https://doi.org/arxiv-2407.19858","url":null,"abstract":"In the field of quantitative finance, machine learning methods have become\u0000essential for alpha generation. This paper presents a pioneering method that\u0000uniquely combines Hidden Markov Models (HMM) and neural networks, creating a\u0000dual-model alpha generation system integrated with Black-Litterman portfolio\u0000optimization. The methodology, implemented on the QuantConnect platform, aims\u0000to predict future price movements and optimize trading strategies.\u0000Specifically, it filters for highly liquid, top-cap energy stocks to ensure\u0000stable and predictable performance while also accounting for broker payments.\u0000QuantConnect was selected because of its robust framework and to guarantee\u0000experimental reproducibility. The algorithm achieved a 31% return between June\u00001, 2023, and January 1, 2024, with a Sharpe ratio of 1.669, demonstrating its\u0000potential. The findings suggest significant improvements in trading strategy\u0000performance through the combined use of the HMM and neural networks. This study\u0000explores the architecture of the algorithm, data pre-processing techniques,\u0000model training procedures, and performance evaluation, highlighting its\u0000practical applicability and effectiveness in real-world trading environments.\u0000The full code and backtesting data are available under the MIT license.","PeriodicalId":501172,"journal":{"name":"arXiv - STAT - Applications","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141866536","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Shotgun DNA sequencing for human identification: Dynamic SNP selection and likelihood ratio calculations accounting for errors","authors":"Mikkel Meyer Andersen, Marie-Louise Kampmann, Alberte Honoré Jepsen, Niels Morling, Poul Svante Eriksen, Claus Børsting, Jeppe Dyrberg Andersen","doi":"arxiv-2407.19761","DOIUrl":"https://doi.org/arxiv-2407.19761","url":null,"abstract":"In forensic genetics, short tandem repeats (STRs) are used for human\u0000identification (HID). Degraded biological trace samples with low amounts of\u0000short DNA fragments (low-quality DNA samples) pose a challenge for STR typing.\u0000Predefined single nucleotide polymorphisms (SNPs) can be amplified on short PCR\u0000fragments and used to generate SNP profiles from low-quality DNA samples.\u0000However, the stochastic results from low-quality DNA samples may result in\u0000frequent locus drop-outs and insufficient numbers of SNP genotypes for\u0000convincing identification of individuals. Shotgun DNA sequencing potentially\u0000analyses all DNA fragments in a sample in contrast to the targeted PCR-based\u0000sequencing methods and may be applied to DNA samples of very low quality, like\u0000heavily compromised crime-scene samples and ancient DNA samples. Here, we\u0000developed a statistical model for shotgun sequencing, sequence alignment, and\u0000genotype calling. Results from replicated shotgun sequencing of buccal swab\u0000(high-quality samples) and hair samples (low-quality samples) were arranged in\u0000a genotype-call confusion matrix to estimate the calling error probability by\u0000maximum likelihood and Bayesian inference. We developed formulas for\u0000calculating the evidential weight as a likelihood ratio (LR) based on data from\u0000dynamically selected SNPs from shotgun DNA sequencing. The method accounts for\u0000potential genotyping errors. Different genotype quality filters may be applied\u0000to account for genotyping errors. An error probability of zero resulted in the\u0000forensically commonly used LR formula. When considering a single SNP marker's\u0000contribution to the LR, error probabilities larger than zero reduced the LR\u0000contribution of matching genotypes and increased the LR in the case of a\u0000mismatch. We developed an open-source R package, wgsLR, which implements the\u0000method, including estimating the calling error probability and calculating LR\u0000values.","PeriodicalId":501172,"journal":{"name":"arXiv - STAT - Applications","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141866623","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An Adaptive Image-denoising Method Based on Jump Regression and Local Clustering","authors":"Subhasish Basak, Partha Sarathi Mukherjee","doi":"arxiv-2407.20210","DOIUrl":"https://doi.org/arxiv-2407.20210","url":null,"abstract":"Image denoising is crucial for reliable image analysis. Researchers from\u0000diverse fields have long worked on this, but we still need better solutions.\u0000This article focuses on efficiently preserving key image features like edges\u0000and structures during denoising. Jump regression analysis is commonly used to\u0000estimate true image intensity amid noise. One approach is adaptive smoothing,\u0000which uses various local neighborhood shapes and sizes based on empirical data,\u0000while another is local pixel clustering to reduce noise while maintaining\u0000important details. This manuscript combines both methods to propose an\u0000integrated denoising technique.","PeriodicalId":501172,"journal":{"name":"arXiv - STAT - Applications","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141866622","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrated Scenario-based Analysis: A data-driven approach to support automated driving systems development and safety evaluation","authors":"Gibran Ali, Kaye Sullivan, Eileen Herbers, Vicki Williams, Dustin Holley, Jacobo Antona-Makoshi, Kevin Kefauver","doi":"arxiv-2407.19975","DOIUrl":"https://doi.org/arxiv-2407.19975","url":null,"abstract":"Several scenario-based frameworks exist to aid in vehicle system development\u0000and safety assurance. However, there is a need for approaches that combine\u0000different types of datasets that offer varying levels of case severity, data\u0000richness, and representativeness. This study presents an integrated\u0000scenario-based analysis approach that encompasses scenario definition, fusion,\u0000parametrization, and test case generation. For this process, ten years of fatal\u0000and non-fatal national crash data from the United States are combined with over\u000034 million miles of naturalistic driving data. An illustrative example\u0000scenario, \"turns at intersection\", is chosen to demonstrate this approach.\u0000First, scenario definitions are established from both record-based and\u0000continuous time series data. Second, a frequency analysis is performed to\u0000understand how often events from the same scenario occur at different\u0000severities across datasets. Third, an analysis is performed to show the key\u0000factors relevant to the scenario and the distribution of various parameters.\u0000Finally, a method to combine both types of data into representative test case\u0000scenarios is presented. These techniques improve scenario representativeness in\u0000two major ways: first, they populate an entire spectrum of cases ranging from\u0000routine events to fatal crashes; and second, they provide context-rich,\u0000multi-year data by combining large-scale national and naturalistic datasets.","PeriodicalId":501172,"journal":{"name":"arXiv - STAT - Applications","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141866535","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Independent fact-checking organizations exhibit a departure from political neutrality","authors":"Sahajpreet Singh, Sarah Masud, Tanmoy Chakraborty","doi":"arxiv-2407.19498","DOIUrl":"https://doi.org/arxiv-2407.19498","url":null,"abstract":"Independent fact-checking organizations have emerged as the crusaders to\u0000debunk fake news. However, they may not always remain neutral, as they can be\u0000selective in the false news they choose to expose and in how they present the\u0000information. They can deviate from neutrality by being selective in what false\u0000news they debunk and how the information is presented. Prompting the now\u0000popular large language model, GPT-3.5, with journalistic frameworks, we\u0000establish a longitudinal measure (2018-2023) for political neutrality that\u0000looks beyond the left-right spectrum. Specified on a range of -1 to 1 (with\u0000zero being absolute neutrality), we establish the extent of negative portrayal\u0000of political entities that makes a difference in the readers' perception in the\u0000USA and India. Here, we observe an average score of -0.17 and -0.24 in the USA\u0000and India, respectively. The findings indicate how seemingly objective\u0000fact-checking can still carry distorted political views, indirectly and subtly\u0000impacting the perception of consumers of the news.","PeriodicalId":501172,"journal":{"name":"arXiv - STAT - Applications","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141866537","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Socio-cognitive Networks between Researchers","authors":"Alejandro Espinosa-Rada, Jürgen Lerner, Cornelius Fritz","doi":"arxiv-2407.21067","DOIUrl":"https://doi.org/arxiv-2407.21067","url":null,"abstract":"Understanding why researchers cite each other has been a longstanding\u0000conjecture in studying scientific networks. Prior research suggests relevance,\u0000group cohesion, or honest source crediting as possible factors. However, the\u0000dual nature of cognitive and social dimensions underlying citation is often\u0000overlooked by not considering the intermediary steps leading up to a citation.\u0000For one work to be cited by another, it must first be published by a set of\u0000authors. Therefore, we investigate the reasons behind researchers' citations,\u0000explicitly examining the interplay of socio-cognitive ties through the\u0000interdependence of coauthorship and citation networks. We assess our claims in\u0000an empirical analysis by employing the Author-Oriented Relational HyperEvent\u0000Model (AuthRHEM) to study Chilean astronomers' citation and collaboration\u0000behavior between 2013 and 2015 in a joint framework. We find evidence that when\u0000deciding which work to cite, authors prefer other work with novelty and\u0000cognitive ties, such as work-to-work relations. At the same time, coherent\u0000groups are relevant because coauthors are cocited more frequently in subsequent\u0000publications.","PeriodicalId":501172,"journal":{"name":"arXiv - STAT - Applications","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2024-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141866620","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}