Articles tagged with Statistical Analysis
Researchers found that selecting the embryo with the lowest predicted risk score reduces disease risk more effectively than eliminating high-scoring embryos. However, polygenic embryo screening offers no guarantees about a baby's health and has limited effectiveness depending on factors like ancestry and age.
Researchers have developed a federated analytics system, FAMHE, that enables healthcare providers to collaborate on statistical analyses and machine learning models without exchanging underlying datasets. The system has been proven mathematically secure and accurately reproduced published studies in multi-centric settings.
Researchers found that males and females have different heat-sensing systems, which may be related to reproduction and offspring care. This difference leads to segregation between the sexes at certain periods during breeding cycles.
A new guideline, AGReMA Statement, provides recommendations for researchers to describe mediation analysis in their papers. The checklist aims to improve the accuracy, completeness, and consistency of reporting mediation analyses.
A study published in PLOS One found that stimulus checks largely reversed post-lockdown spending declines and stimulated spending among people making less than $35,977 per year. The research analyzed 5.6 million daily transactions and identified local 'hot spots' of spending changes.
The CAPTIVA trial will enroll 1,683 patients with symptomatic intracranial arterial stenosis and determine if two new prospective therapies are more effective than aspirin plus clopidogrel at reducing the risk of a second stroke. Intensive medical management and lifestyle modifications will also be provided to all patients.
A new study found that local mobility, primary care access, and limited contacts in contagion hubs can curb COVID-19 deaths. The researchers used novel statistical approaches to analyze mortality curves across 20 Italian regions, identifying a strong association between mobility and COVID mortality.
The study found that patients with low risk of exacerbations do not benefit from low to moderate dose inhaled corticosteroids, while high doses are associated with severe side effects. Tailoring treatment according to patient characteristics, such as age and blood cell count, can minimize over- and underuse of inhaled corticosteroids.
Researchers analyzed social media connections of over 4 million users across 10 countries, finding common characteristics and behaviors, including follow ratios and profile length. These findings can help tailor data analysis to cultural differences, improving marketing and information sharing.
A new study estimates the intensity and frequency of extreme novel pandemics, finding that large pandemics like COVID-19 are relatively likely due to growing risks. The study suggests that a pandemic similar in scale to COVID-19 is likely within 59 years, highlighting the need for adjusted perceptions of pandemic risks.
Research in the southern Murray-Darling Basin has found a downward spiral in water availability, driven by accelerated changes in climate. Since the 1990s, cool season rainfall and river heights have significantly decreased due to increased evaporation and changed precipitation patterns.
A big data study from UNSW Sydney found that Australian cancer patients kept up their pharmaceutical treatments during last year's COVID-19 lockdowns. The researchers attribute the good news to relatively low rates of COVID-19 infections in Australia, which minimally impacted cancer treatment patterns.
Researchers analyzed inner ear anatomy in extinct and living species to understand dinosaur sensory biology and behavior. The studies found clear patterns relating the shape of the semi-circular canal and cochlea to movement ability and hearing high-frequency sound.
Researchers analyzed California wildfire data from 2000-2019, finding the annual burn season has lengthened and the yearly peak shifted from August to July, fueled by climate change and human-caused ignitions. Small fires account for most incidents, but large fires have deadly consequences.
University of Groningen researchers used artificial intelligence to analyze the Dead Sea Scrolls, identifying two scribes who wrote the famous Great Isaiah Scroll. The AI tool separated text from background and analyzed handwriting patterns, revealing differences between the two scribes.
A state-by-state analysis of mask-wearing policies and COVID-19 rates reveals a strong association between high adherence to mask wearing and reduced spread of the virus. States with high mask-wearing rates had significantly lower COVID-19 rates, highlighting the importance of this public health measure in mitigating the pandemic.
Nilabja Guha, a UMass Lowell professor, received a three-year NSF grant to develop statistical tools for analyzing changes in data. The project aims to fill a gap in the analysis of changes in data and will focus on Bayesian change-point analysis.
Researchers found most British legal documents from this period were written on sheepskins, which allowed for easy detection of fraudulent changes. The high fat content of sheepskins made it difficult to alter text without leaving a visible mark.
The study analyzed subtitles of 2,000 films from the past seven decades to find consistent biases in Bollywood and Hollywood movies. Bollywood films consistently depicted fair skin as a beauty standard, while male characters made up about 60-65% of all pronouns used.
Analysis of 33,000 graves reveals a change in Western European burial practices from regionally specific grave goods to standardized, unfurnished burials between the 6th-8th centuries AD. This shift indicates a well-connected Europe with regular contact and exchange of ideas across vast areas.
Physicist David Wolpert developed a hybrid formalism to analyze far-from-equilibrium systems with explicitly distinguished subsystems. The formalism combines advances in nonequilibrium statistical physics and graphical models, allowing for the study of nanoscale systems like biological cells.
Researchers have developed a new statistical model that accurately analyzes electrodermal activity (EDA) signals, providing insights into internal cognitive states and physiological processes. The inverse Gaussian model matches the physiology of sweat production and offers a principled approach to modeling EDA signals.
Research shows that appointing a new leader just before an election leads to a higher turnover of MPs across Europe. The study analyzed statistics from 251 parties between 1945 and 2015, finding that the proportion of new and re-elected MPs is almost 5 percentage points higher for parties with a new leadership.
Most of Stonehenge's sarsen megaliths originate from West Woods, Wiltshire, according to a new analysis. The study used advanced techniques to determine the stones' chemical composition, revealing a common origin for the massive boulders.
A new study identifies weak concrete and heavy trucks as the top reasons for sewer line failure. The researchers used statistical analysis to show that these factors are most likely to trigger a problem among common causes of sewer pipe failure.
A Bayesian statistical analysis of life's emergence and development on Earth predicts that if history were repeated, life would likely emerge similarly early. However, the emergence of intelligence might not guarantee a repeat occurrence due to its rarity.
A new statistical method called distance multivariance can detect higher-order dependencies between multiple variables, even complex non-linear relationships. This method has been applied to a survey of school students, revealing a higher-order dependence including age and gender as influencing factors for missed school days.
Researchers are using RPPA technology to analyze ischemic stroke specimens, focusing on protein content and identification of specific proteins of interest. The project aims to demonstrate the utility of this approach in understanding thrombus formation.
Researchers analyzed over 4,000 Instagram photos from various cities worldwide and found that people tend to centre one of their eyes slightly to the left of centre of the selfie, with the left eye being more commonly centred than the right. This phenomenon is consistent with a previously observed cognitive bias known as pseudoneglect.
Researchers at the University of Waterloo found that speeding is a strong predictor of car crashes, while hard braking and acceleration had no statistically significant links. The analysis used data from 28 million trips and suggests that insurance companies should look at speed when assessing driver risk.
Researchers at Queensland University of Technology (QUT) have developed a software tool that can predict an athlete's level of fatigue and its impact on their performance. The tool uses statistical analysis to analyze data from countermovement jump tests, providing insights into metabolic and neuromuscular fatigue.
A new statistical analysis published in the Bulletin of the Seismological Society of America has found no correlation between deep-sea fish appearances and earthquakes in Japan. Despite long-held folklore suggesting otherwise, researchers have only identified one potentially correlated event out of 336 sightings and 221 earthquakes.
Death rates from falls for US adults 75 or older rose significantly from 2000 to 2016, with a 128% increase in absolute deaths and a 131% increase in death rates per 100,000 people. This trend was observed across different age groups, particularly among men aged 75-79 and women aged 95 or older.
A new study published in Science Advances suggests that the South African fossil Australopithecus sediba is unlikely to be the direct ancestor of humans. Instead, researchers believe that Australopithecus afarensis, known for the 'Lucy' skeleton, is a stronger candidate.
A study analyzing nearly 57,000 US military casualties found that the use of tourniquets, blood transfusions, and faster prehospital transport significantly reduced mortality. These interventions accounted for a large part of the increase in survival rates among critically injured soldiers.
Researchers analyzed 2,400 languages to understand how human diets impact consonant prevalence. A biomechanical model revealed that a shift from hunter-gatherer to agricultural societies led to changes in bite configuration and labiodental sound distribution.
A study published in JAMA Network Open reveals a slight increase in readmission rates among patients discharged from Swiss hospitals under flat-rate payment structures. The researchers found that the rate of 30-day readmissions rose from 14.4% to 15.0%, raising concerns about early discharges and potential impact on hospital quality.
Researchers analyzed data from professional sports leagues and online games to find that past shared success among team members boosts their chances of winning future games. The study's findings suggest that team chemistry is a crucial factor in determining team success, transcending cultural differences and industries.
A new study by San Francisco State University researchers found that offering job-related support and emotional support have roughly the same effect. Simply making resources available is often more consistently positive than overtly discussing it. Employers are advised to think carefully before providing support, considering whether it...
The article discusses the complexity of data ethics, emphasizing personal data ownership, consent, trustworthiness, and privacy. Researchers aim to strike a balance between harnessing data's potential benefits and mitigating its risks.
Researchers analyzed fossilized backbones and reconstructions to better understand the evolution of mammals' distinct spinal regions. The study found that forelimb reorganization played a crucial role in driving regionalization across the spine, leading to the varied forms of modern mammals.
The Analytics and Data Science Institute created nine research laboratories to tackle business and public service data problems. Students from various backgrounds collaborate to understand statistical and computer science perspectives, enabling them to communicate effectively with non-experts.
Researchers used data analysis techniques, including decomposing songs into musical features, to determine Beatles songwriting styles. The approach identified distinct patterns in Lennon and McCartney's compositions, allowing for the prediction of disputed authorship.
A recent study analyzed NFL injuries from 2007-2016 and found that the increase in injuries occurred faster after practice restrictions took effect, not as a result of them. The study also showed no significant differences in injury rates for different muscle groups or types of injuries.
Home run numbers increased significantly between 2016-2017 due to a drop in vertical break of pitches, suggesting pitchers aimed to reduce home runs. However, this strategy failed to work as other factors contributed more to the surge.
A team of scientists developed a mathematical model to analyze antibody levels in Thai patients, identifying a 'window of risk' for individuals who are more susceptible to severe forms of the disease. The study's findings offer new insights into dengue infection and may lead to closer monitoring and vaccination strategies.
Researchers have identified three aldehydes in the skin of infected children that attract mosquitoes, which could lead to better malaria detection and control. The study's findings provide insight into the parasite's infection route and offer opportunities for developing new biomarkers and insect traps.
A new study reveals hostile neighbors use subversion to keep weak countries weak, impacting security, economic growth and human well-being. Lead author Melissa M. Lee's research found that bordering states or outside actors significantly affect a country's ability to govern itself.
A new evaluation method for AI predictability has been developed by Kyoto University researchers, tackling issues of accuracy and bias in traditional metrics.
Researchers analyze order flow for pairs of NASDAQ 100 stocks and find four groups with large mutual differences, surprising given the lack of reflection in actual prices. This discovery contributes to modeling price evolution and could be used to evaluate the impact of financial crises.
Researchers validated a new earthquake forecasting model based on the Amatrice-Norcia sequence, showing improved accuracy compared to existing models. The 'brick-by-brick' approach of the operational earthquake forecasting system may revolutionize seismologists' ability to forecast earthquakes.
Researchers found 27 distinct dimensions of emotion that account for hundreds of people's reported feelings in response to video clips. The study challenges the notion that emotions are finite clusters and instead reveals smooth gradients between emotions.
Researchers at MIT and Cisco Systems have developed a new approach to network monitoring called Marple, which allows routers to report on their own state without relying on external servers. By using a programming language and simple circuit design, Marple enables precise traffic analysis while reducing the need for additional hardware.
Researchers analyzed climatological and sorghum yield data from Kansas, finding that temperatures above 33°C result in declining yields. Moderate and high temperature increases of 2°C and 4°C led to significant yield losses, with an average reduction of 10% per 1°C from 1-5°C.
Advanced statistical modeling tools can analyze large electronic health records to inform medical decisions and develop individualized treatments for high-risk patients. The methods leverage machine learning and flexible models to draw valid inference from representative populations.
A recent statistical analysis suggests that the changed strike zone could negatively affect batter productivity. The study found pitch location-specific differences in run expectancies, which may lead to internal struggles among players and impact salaries.
The Well-Being Map is an interactive tool that uses over 37 billion publicly shared tweets to analyze and display scores for various characteristics in every US county. It provides valuable insights into community traits such as well-being, personality, government-reported health, and socioeconomic factors.
A new study by University of Waterloo researchers found that smart meters and time-of-use pricing reduced residential energy demand during peak periods by only 2.6% and 2.4%, respectively. The modest changes in behavior raise concerns about whether the cost of installing smart meters has been justified.
A new analysis of Medicare's Hospital Compare portal found that the statistical methodology used to rate and compare hospitals underestimates mortality rates of acute myocardial infarction (AMI) at small hospitals. The research proposes an alternative model that incorporates additional factors such as hospital volume, nursing staff, an...
Industry experts examine the advantages of shifting data analytics to the cloud, including cost savings and improved management of analytics workloads. The panel discussion highlights the potential for companies to gain by moving their data analytics activities to the cloud, including more efficient use of IT resources.