Articles tagged with Statistical Methods
Statisticians combine academic and commercial techniques to analyze large data sets, reducing processing time and increasing insight. This technology has the potential to aid researchers in various fields, including astronomy, by providing quicker insights from Big Data.
Researchers developed a 'reusable holdout' tool to identify meaningful correlations in adaptive data analysis. The algorithm, tested on two rigged datasets, correctly identified true statistical significance and eliminated false discoveries.
The center, led by Iowa State University, aims to develop tools for evaluating pattern and digital evidence analysis methods and an education program for judges, lawyers, and forensic science investigators. This will enhance the scientific basis for forensic evidence used in the criminal justice system.
A recent study using a new statistical methodology found that the patient mortality rates of new and experienced surgeons are similar. However, researchers emphasize that further studies are necessary to confirm these findings and account for other factors, such as changes in surgical training practices and patient demographics.
Nate Silver will receive the 2015 JPBM Communications Award for his innovative use of statistics and probability to make accurate predictions in various fields, including politics and baseball. The award recognizes his ability to communicate complex mathematical concepts to a broad audience.
Brown University professors Joseph Hogan and Rami Kantor will use a new $3.5 million grant to develop data-driven methods for improving HIV treatment monitoring decisions, potentially reducing costs and increasing effectiveness in low- and middle-income countries.
The new journal ASCMO addresses the need for an outlet to publish statistical methodology work in climate change and atmospheric/oceanic applications. It will feature accessible articles on scientific advances and statistical methods for interdisciplinary researchers.
A new study published in Language presents a rigorous linguistic analysis that contradicts previous research on the origins of written language. The analysis suggests that popular statistical measures do not support the claim that certain symbol systems, such as the Indus Valley and Pictish symbols, are examples of written language.
A new measurement model reveals that human rights are actually improving, contrary to common claims. The model takes into account changes in data collection and categorization, showing more countries with better human rights records.
A new study reveals that the South Asian summer monsoon is experiencing significant changes in extreme wet and dry events, leading to an increased risk of drought and flood in central India. The researchers found that the intensity of wet spells and the number of dry spells have both been increasing in recent decades.
Scientists have created a new statistical method to analyze individual cells, allowing for more accurate detection of cell properties. By combining data from multiple cells, researchers can reduce errors and overcome statistical noise, leading to better understanding of cellular behavior and development.
US researchers found that graphic warning labels similar to those in Canada would reduce smoking rates by 2.9-4.7 percentage points, resulting in 5.3-8.6 million fewer smokers in the US. The study challenges the FDA's regulatory impact analysis and calls for a revised proposal to require graphic warnings.
Seven Texas A&M faculty members have been recognized by the American Association for the Advancement of Science (AAAS) for their contributions to various fields. They include Nancy M. Amato, Perla Beatriz Balbuena, Raymond J. Carroll, Jonathan C. Coopersmith, Bani K. Mallick, H. Joseph Newton, and Matthew S. Sachs.
Researchers developed a statistical modeling approach to assemble and evaluate the fitness of individual neurons, identifying diverse yet robust populations as the most successful. This method can help understand the importance of neuronal diversity and predict how alterations in neuron variability impact function.
Researchers at Max-Planck-Gesellschaft developed a new keyboard layout called KALQ that minimizes thumb movement and places frequently used letters centrally. Trained users can achieve 37 words per minute, significantly faster than the 20 words on split QWERTY, with improved error correction methods
Researchers at UMass Amherst have developed a new method called MixMAP that uses existing public databases to reveal substantial new information about genes associated with high cholesterol and heart disease. The approach provides opportunities for developing new treatments and identifying people at greatest risk.
A team of researchers has found significant errors in a previously published study claiming to identify awareness in vegetative state patients using bedside EEG technology. The reanalysis questions the validity of the original study's statistical methodology and findings.
Researchers found that comatose patients' brain networks are restructured in a radical way, with some regions becoming hubs and others less connected. This changes could be linked to consciousness disorders and may help clinical practitioners diagnose comatose patients more accurately.
A new study tested Cope's Rule, which suggests that small animals evolve into bigger ones, on non-avian dinosaurs. The research found that some groups of dinosaurs, such as ceratopsids and hadrosaurs, tend to increase in size over time, while others, like four-legged sauropods, show no upper limit to their body size.
Researchers found long-range correlation between letters and words within texts, enabling conclusions about topic connections. Burstiness analysis also reveals representative words in passages.
Researchers at Tel Aviv University discovered a mathematical system used by ancient Egyptians to measure the volume of spherical jugs. By linking units of length to units of volume, merchants could precisely determine the quantity of liquid within, allowing for accurate pricing and trade.
Researchers discovered that random patches of disordered electric charges can induce a twisting force strong enough to affect biological objects at nanometers or micrometers away. This phenomenon could help understand patterns in biology, such as lock and key interactions.
A new statistical model developed by Penn State researcher Cedric Neumann allows for the quantification of fingerprint evidence, paving the way for its full inclusion in criminal identification processes. The model enables the evaluation of fingerprint evidence on a continuous scale, rather than relying on subjective expert opinions.
A team of MIT cognitive scientists found that ambiguity in language allows for the reuse of short, efficient sounds that listeners can easily disambiguate with context. This theory contradicts previous arguments that ambiguity is a problem for communication, suggesting it's actually a key to making language more efficient.
A new mathematical model can detect financial bubbles in real-time using tick data and statistical techniques. The model estimates asset price volatility to determine if a bubble is forming, which can help prevent economic collapses.
Researchers at Ruhr-University Bochum have developed a new method to measure the distance between stationary and non-stationary processes, enabling more accurate statistical tests. This approach makes it possible to construct better models for predicting extreme events in time series data.
Researchers are developing new data analysis and modeling methodologies to aid the transition from the decennial census to the annual American Community Survey. These methods could impact how over $400 billion in federal and state funds are distributed annually.
Researchers found significant biodiversity loss in streams with less than 2% developed land, contrary to previous estimates. Chemical pollution is implied as a major contributor to the degradation of water quality.
Researchers used multidimensional scaling to create a graphical plot of amino acid sequence data for all strains of H3N2, predicting the dominant strain with high accuracy. The new method can predict whether a new flu strain will become dominant as little as two weeks after its appearance in GenBank, outperforming current methods.
Researchers found that screening for lung cancer using helical CT scanning reduced lung cancer deaths by 20 percent compared to chest X-rays. The study also showed overall deaths due to any factor were 7 percent lower in those screened with low-dose helical CT.
A pilot study in North Carolina found that mailing surveys instead of calling people on the phone yields higher response rates and more complete data. This method is being tested as a replacement for the current home telephone survey used by NOAA to determine which areas off the coast are overfished.
Bernd Sturmfels received the John von Neumann Lecture for his work on applying algebraic ideas to problems in biology, statistics, optimization, and polynomial systems. The lecture surveyed recent applications of abstract algebra in modeling and solving nonlinear problems.
Researchers at NIST have designed an instrument that can quickly and precisely detect trace amounts of chemical compounds indicating poultry spoilage. The technique relies on identifying low-volatility compounds present early in the decay process.
A new study by UT Southwestern Medical Center suggests that appendicitis may be caused by undetermined viral infection or infections. The researchers evaluated data over a 36-year period and found seasonal variations and clustering of cases, supporting the theory that appendicitis may be a viral disease.
The PLoS Medicine editors discuss maintaining integrity in clinical effectiveness research, highlighting the need for transparent methods and objective reporting. They also warn against practices that distort scientific evidence, such as cherry-picking studies or fishing for favorable conclusions.
The National Institute of Statistical Sciences (NISS) has received funding to develop Bayesian methods for syndromic surveillance using conditionally auto regressive models. The research aims to provide quantified estimates of disease presence, characterize uncertainties, and implement computational methods at a nationwide scale.
A recent study by Penn State researchers suggests that several statistical methods used to detect natural selection at the molecular level are faulty. This has implications for the interpretation of hundreds of published studies on human evolution and the role of natural selection in shaping human traits.
A report by Clara Fang and colleagues analyzed crash data before and after implementing traffic calming measures in Hartford. The study found that all safety features led to fewer crashes, with greater reductions on streets with higher initial crash rates.
A Montana State University study has found that the coldest daily temperatures recorded in Bozeman, Mont., and Coldstream, British Columbia, have occurred less often over the past several decades. Extreme warm nighttime temperatures have become more frequent at a rate of about one percent per year or 10 percent per decade.
A researcher at the University of Texas at Austin has identified a statistical method to select people for secondary security screenings more efficiently and fairly. Square root bias sampling can reduce resource waste by screening individuals from higher-risk groups less frequently while still prioritizing them.
The study aims to separate dysfunctional brain activity from 'noise' using spatial statistical modeling, a new technique that uses neighboring locations to strengthen weak signals. The research could help identify specific areas of the brain affected by the syndrome in Persian Gulf War veterans.
A MU researcher found that while traditional and blended learning environments have few statistical differences in effectiveness, students in hybrid classrooms are more frequently confused about course requirements. The study suggests that motivation and technological ability play a major role in success in blended environments.
Roger Schroeder examines Six Sigma's effectiveness in cutting costs and improving profitability. He defines the methodology and builds an academic framework, identifying consistent elements such as parallel-meso structures, strategic project selection, leadership engagement, and structured methodologies.
Researchers in Spain have compiled 20th century temperature data to show evidence of climatic change. They found signs of 'pollution' that needed to be removed through homogeneization, a process that ensures the data reflects only climatic evolution.
Researchers at Uppsala University found that protein signal transfer may occur within small proteins, contradicting a previous study using statistical methods. Experiments showed nearby parts of the protein interact more than distant parts.
A study published in BMC Public Health found a 42% reduction in suicide rates among 15-29 year old men in Scotland between 2000 and 2004. The decline is attributed to a decrease in hangings as a means of suicide, with other methods such as poisoning becoming more prevalent.
Researchers have developed a new method for presenting clinical trial survival data that includes all participants, overcoming the limitations of traditional Kaplan-Meier plots. This new approach provides a more accurate representation of treatment effects, enabling physicians and patients to better understand clinical trial results.
Researchers found that different statistical methods can produce varying results when analyzing observational data. Using instrumental variable analysis, they found a 16% relative decrease in death rate, whereas other methods showed a 50% relative decrease.
Researchers develop new statistical approaches to assist doctors' decision making for severely ill newborn babies, enabling better-informed decisions and potential life-saving interventions. Bayesian methods will be integrated into electronic medical records to provide instant analysis of patient data and improve treatment strategies.
St. Jude Children's Research Hospital has developed a new statistical technique that enables researchers to statistically analyze results of clinical trials without traditional control groups. This novel approach allows for the early adoption of new treatments and validates the use of interim analyses in clinical trials.
Research comparing different screening instruments found that women prefer self-completed questionnaires over face-to-face interviews for disclosing experiences of intimate partner violence. The written WAST format yielded significantly less missing data, making it a suitable option for both clinical and research applications.
A USC Information Sciences Institute system identifies alpha chatterers in online communities, enabling the extraction of statistically correct answers. The method integrates speech act analysis, lexical similarity, and poster trustworthiness to create links for interpretation.
A statistical method developed by NIST makes analyzing satellite-borne time signals faster and easier, enabling accurate location and distance measurements. The method also accelerates the evaluation of clocks during satellite construction, improving GPS reliability and accuracy.
Researchers have developed a new statistical method to identify linked genomic loci influencing gene expression in yeast, revealing 37% of gene expression traits link to two loci. The technique bypasses overwhelming computations and provides insights into the genetic basis of complex traits.
A new method of controlling blood levels of calcineurin inhibitors (CNIs) improves the success of liver transplants by reducing variation and toxicity. The four-step process increases CNI levels within target ranges from 50% to 80% in patients.
A new tool for tracing human gene connections has been developed, enabling the study of complex systems like power grids and neural networks. The research used a novel approach to analyze gene activity in response to c-Myc activation, revealing over 1,000 genes significantly altered by the protein.
Virtual colonoscopy is effective in detecting large colorectal polyps and masses, but is less sensitive for smaller ones. The review found no direct evidence on the effect of screening methods on cancer incidence or death prevention.
According to a recent study, youth drinking rates have stabilized in the US since the early 1990s, with over 80% of adolescents consuming alcohol by age 12th-grader. The stable prevalence rates are concerning, highlighting the need for continued efforts to prevent underage drinking.
A study published in BMC Medical Research Methodology revealed that over 38% of Nature papers and a quarter of BMJ articles contained at least one statistical error. The researchers suggest making raw data freely available to minimize the impact of errors, which could have important practical consequences.
A study by Wildlife Conservation Society argues that India's tiger counting method using pugmarks is flawed and leads to poor conservation. The authors suggest adopting simpler, statistically sound monitoring methods such as sample surveys and encounter-rate surveys.