Tip sheet for joint statistical conferences Aug. 2 – 6, 2020

WHAT

The 2020 Joint Statistical Meetings will deliver collectively statisticians and information scientists from all over the world – this yr, for the primary time, in a digital format. This tip sheet from the American Statistical Association highlights fascinating shows from the upcoming JSM 2020.

WHEN

Sunday, August 2 – Thursday, August 6, 2020

WHERE

Online venue https://ww2.amstat.org/meetings/jsm/2020/

Complementary press registration is open, courtesy of the ASA. Information on find out how to obtain press credentials is on the backside of this launch.

FEATURED RESEARCH (SYNOPSES BELOW)

**ASTERISKS IDENTIFY CORONAVIRUS/COVID-19-RELEVANT RESEARCH

MONDAY HIGHLIGHTS

1. Improving NCAA soccer rankings with information science

2. Precision drugs for stem cell transplants

3. Statistical evaluation of footprints in forensic science

4. The function of unsure an infection standing in controlling epidemics **

5. Deep Learning AI neural networks for local weather change

TUESDAY HIGHLIGHTS

6. Which emoticon persona are you? 😉

7. Undercounting invisible immigrant communities within the census

8. How do algorithmic instruments have an effect on equity and high quality of determination making?

9. Mathematical mannequin for re-opening companies throughout COVID-19 **

WEDNESDAY HIGHLIGHTS

10. Analyzing the equity of a pre-trial danger algorithm

11. Data science instruments for monitoring affected person security throughout medical trials

12. “Nowcasting” and Forecasting COVID-19 **

13. Text-mining information articles to foretell inventory returns

THURSDAY HIGHLIGHTS

14. Group testing for COVID-19: What is it, and the way may it’s applied extra successfully? **

15. Panel on statistical significance and P-values

16. COVID-19 Infectious illness modeling and statistics: myths, maxims and mobilization **

17. New mannequin for contact tracing and illness unfold **

18. Who are the scientific grant gatekeepers?

19. Statistical fashions for evaluating state opioid insurance policies

MONDAY HIGHLIGHTS

1) Improving NCAA soccer rankings with information science

In US school soccer, declaring a nationwide champion hasn’t been simple. Prior to 2014, the statistical score technique used was plagued with criticisms, and presently a 13-member committee selects and seeds groups for playoffs. But some followers nonetheless marvel if there’s a greater manner. In this presentation, Shane Reese of Brigham Young University will current a brand new statistical score system, referred to as Ratings Using Score Histories, developed with colleagues to assist choose playoff groups. Its novel characteristic: it makes use of information from a recreation’s rating course of – that’s, the rating for every time limit all through a recreation – for a complete season. Unlike earlier strategies, this new system treats groups from weaker and stronger conferences extra pretty and likewise makes use of all accessible information. The presentation will show how the score system can be utilized, together with outcomes from the 2019-2020 season.

The presentation, “RUSH: An Evolutionary Approach to Ranking College Football Teams,” will happen Monday, August 3, 2020 : 10:00 AM to 11:50 AM EDT. Abstract: https://ww2.amstat.org/meetings/jsm/2020/onlineprogram/AbstractDetails.cfm?abstractid=309359

2) Precision drugs for stem cell transplants

Treating blood cancers like leukemia generally requires the aggressive method of transplanting stem cells from a wholesome donor right into a affected person’s bone marrow. To assist put together the affected person’s physique for the transplant, a drug referred to as busulfan is injected immediately into the veins. It’s tough to get its precise dosage appropriate, nevertheless – an excessive amount of can result in toxicity and even demise, whereas too little could make it simpler for the most cancers to return. In this presentation, Peter Thall from the MD Anderson Cancer Center on the University of Texas will describe the brand new “precision medicine” statistical mannequin that he and colleagues created to find out the best dosage, leading to a way that be simply utilized by any transplant physician. By switching from the present “one-size-fits-all” technique to the brand new technique, the researchers calculate that docs can prolong many sufferers’ lives dramatically – by a median of 10 to 14 months, for instance, for 40 to 60-year-olds in full remission, which is an enchancment of as much as 290%.

The presentation, “Bayesian Nonparametric Survival Regression for Optimizing Precision Dosing of Intravenous Busulfan in Allogeneic Stem Cell Transplantation,” will happen Monday, August 3, 2020 : 10:00 AM to 11:50 AM EDT. Abstract: https://ww2.amstat.org/meetings/jsm/2020/onlineprogram/AbstractDetails.cfm?abstractid=308042

3) Statistical evaluation of footprints in forensic science

Using footprints in forensic science to hyperlink a suspect to a criminal offense scene is statistically sophisticated. All shoe soles have some type of random markings – stray holes and scratches acquired throughout regular use – which investigators can use to match a suspect’s shoe with a criminal offense scene print. But first investigators have to mathematically perceive how these random markings accumulate within the first place. In this presentation, Naomi Kaplan-Damary on the University of California, Irvine and the Hebrew University of Jerusalem will current work with colleagues that concerned the evaluation of almost 400 sneakers. Using the identical equations that may describe the distribution of bushes in a forest or stars within the Milky Way, the researchers pinpointed the areas on the soles which might be extra prone to choose up distinguishing blemishes, which can assist decide their significance as proof.

The presentation, “A Step Forward in Estimating the Probability of Accidental Mark Locations on a Shoe Sole,” will happen Monday, August 3, 2020 : 1:00 PM to 2:50 PM EDT. Abstract: https://ww2.amstat.org/meetings/jsm/2020/onlineprogram/AbstractDetails.cfm?abstractid=309225

4) The function of unsure an infection standing in controlling epidemics **

During an epidemic, public well being officers ideally would know precisely who was contaminated at any second so they might shortly intervene to cease the unfold (by quarantining the contaminated people, for instance). In actual life, in fact, there’s normally loads of uncertainty round who’s contaminated and who’s not. This uncertainty implies that some uninfected individuals will likely be needlessly quarantined, disrupting households and workplaces, whereas some contaminated individuals will likely be allowed to combine freely and unfold the illness sooner. Jessica Hoffman from University of Texas at Austin will current theoretical outcomes with colleagues displaying that even a tiny uncertainty has a dramatic affect on the period of time and sources (resembling quarantining) wanted to include the epidemic. This work implies {that a} group ought to spend money on realizing precisely who’s contaminated – by contact tracing, for instance – or else it’ll pay the value tenfold later.

The presentation, “The Cost of Uncertainty in Curing Epidemics,” will happen Monday, August 3, 2020 : 1:00 PM to 2:50 PM EDT. Abstract: https://ww2.amstat.org/meetings/jsm/2020/onlineprogram/AbstractDetails.cfm?abstractid=312849

5) Deep Learning AI neural networks for local weather change

Deep Learning is a robust kind of machine studying synthetic intelligence that has been used with extraordinary success in fields resembling laptop imaginative and prescient, speech recognition, and language translation. But its use in local weather change science is new. Prabhat from Lawrence Berkeley National Laboratory and colleagues have labored on utilizing Deep Learning for the issue of detecting excessive climate occasions resembling hurricanes and extreme climate fronts. In this presentation, Prabhat will describe their work coaching a state-of-the-art deep studying community to seek out excessive climate patterns in advanced “ground-truth” local weather datasets. He may even present how they will now apply the skilled community to new local weather datasets and use this to know how excessive climate patterns will change sooner or later.

The presentation, “Deep Learning for Extreme Weather Detection,” will happen Monday, August 3, 2020 : 1:00 PM to 2:50 PM EDT. Abstract: https://ww2.amstat.org/meetings/jsm/2020/onlineprogram/AbstractDetails.cfm?abstractid=309281

TUESDAY HIGHLIGHTS

6) Which emoticon persona are you? 😉

Before emojis there have been emoticons – primitive emoji pictographs that conveyed emotion by plain-text characters. Users may select between totally different types – a smiley of 🙂 or :D, for instance, as an alternative of :). Do individuals select totally different emoticon types at random, or do patterns exist that reveal one thing about an individual? To examine, Juha Alho from the University of Helsinki analyzed discussions between 2001 and 2015 on suomi24, a big Reddit-style social-networking website in Finland, for a complete of 48 million particular person posts. Using a statistical technique referred to as correspondence evaluation, Alho uncovered 4 distinct emoticon consumer “personalities”: the Classics, the Noses, the D-Grins, and the Multi-Mouths. In this speak Alho will describe these persona teams, talk about each day emoticon utilization patterns, present developments over time, and reveal which sports activities boards – from golf to ice swimming to parkour – had the biggest relative shares of every of the 4 emoticon personalities.

The presentation, “What Authors Reveal of Themselves in Internet Discussions?,” will happen Tuesday, August 4, 2020 : 10:00 AM to 11:50 AM EDT. Abstract: https://ww2.amstat.org/meetings/jsm/2020/onlineprogram/AbstractDetails.cfm?abstractid=308134

7) Undercounting invisible immigrant communities within the census

It’s been estimated that nearly 11 million undocumented immigrants presently stay within the US. This quantity, and their potential exclusion on the census, has grow to be a hotly debated political situation. How do researchers even estimate the variety of undocumented immigrants, and what occurs if they don’t seem to be counted? In this presentation, Nadia Flores-Yeffal of Texas Tech University will reply these questions and supply context across the counting of invisible communities of immigrants. She estimates that the undercount within the 2020 Census of each undocumented immigrants and their US-born members of the family may very well be as much as 8% of your complete inhabitants, due partially to the prevalence of “mixed-status families.”

The presentation, “How Are Invisible Communities of Immigrants in the United States Counted? What Happens If They’re Undercounted?,” will happen Tuesday, August 4, 2020 : 10:00 AM to 11:50 AM EDT. Abstract: https://ww2.amstat.org/meetings/jsm/2020/onlineprogram/AbstractDetails.cfm?abstractid=312361

8) How do algorithmic instruments have an effect on equity and high quality of determination making?

In right now’s data-rich society, lots of our choices are guided partly by machine suggestions, from on-line procuring to film suggestions. Judges typically use algorithm suggestions as effectively, for instance when weighing the dangers of releasing an arrestee on bail earlier than a trial. Much of the talk round these pretrial risk-assessment devices has centered on accuracy and equity of the algorithms themselves, nevertheless, not on how the algorithms affect and form their customers’ habits. Kosuke Imai from Harvard University and colleagues have developed a statistical framework for experimentally evaluating the affect of machine suggestions on human choices, together with whether or not or not they enhance the equity of choices or result in choices with higher outcomes. This presentation will illustrate the brand new strategies with an instance from the felony justice system, displaying how using a risk-assessment algorithm influenced judges’ choices and whether or not it resulted in racial or gender bias in outcomes.

The presentation, “Experimental Evaluation of Computer-Assisted Human Decision Making: Application to Pretrial Risk Assessment Instrument,” will happen Tuesday, August 4, 2020 : 1:00 PM to 2:50 PM EDT. Abstract: https://ww2.amstat.org/meetings/jsm/2020/onlineprogram/AbstractDetails.cfm?abstractid=309356

9) Mathematical mannequin for re-opening companies throughout COVID-19 **

How can leaders resolve find out how to reopen the economic system safely whereas the coronavirus continues to be circulating within the inhabitants? In this presentation, Hongyu Miao of the University of Texas Health Science Center at Houston will describe work with colleagues in creating a mathematical mannequin that considers each revenue and an infection danger from a enterprise entity’s perspective. They suggest an algebraic equation that describes the online revenue a enterprise can generate by reopening and likewise shouldering the prices related to virus suppression and employee safety. The presentation will illustrate the mannequin with case research, talk about what function private protecting gear ought to play within the office, and present how a enterprise may management an infection charges in a office whereas additionally producing a constructive web revenue.

The presentation, “Modeling of Business Reopening when Facing SARS-CoV-2 Pandemic: Protection, Cost and Risk,” will happen Tuesday, August 4, 2020 : 1:00 PM to 2:50 PM EDT. Abstract: https://ww2.amstat.org/meetings/jsm/2020/onlineprogram/AbstractDetails.cfm?abstractid=313328

WEDNESDAY HIGHLIGHTS

10) Analyzing the equity of a pre-trial danger algorithm

Judges typically use algorithms to assist categorize the chance that an arrested particular person will commit one other crime earlier than their trial or fail to seem for court docket dates. The equity of those of those pre-trial danger evaluation instruments has been referred to as into query in recent times, nevertheless. In this presentation, Megan Price from the Human Rights Data Analysis Group will current work with colleagues that investigated the equity of such a instrument in San Francisco. Their work regarded on the algorithm’s sensitivity to “overbooking,” the place a defendant is booked on extra severe prices which might be finally dropped, generally in alternate for a responsible plea to lesser prices. Their outcomes confirmed that in additional than 1 / 4 of the instances, overbooking was related to the defendants receiving stricter pre-trial suggestions than they might have acquired in any other case. The researchers say this raises questions concerning the appropriateness of those instruments for high-stakes conditions.

The presentation, “Assessing Risk Assessment in San Francisco,” will happen Wednesday, August 5, 2020 : 10:00 AM to 11:50 AM EDT. Abstract: https://ww2.amstat.org/meetings/jsm/2020/onlineprogram/AbstractDetails.cfm?abstractid=309351

11) Data science instruments for monitoring affected person security throughout medical trials

Clinical trials that research the effectiveness of recent medication – together with potential COVID-19 therapies – have to preserve cautious watch on a large number of affected person well being measurements to ensure the routine is secure. Traditionally, this information has been reported to clinicians in pages of static tables and lists, which makes it arduous to identify vital patterns. In this presentation, James Buchanan from Covilance and his colleagues from a multidisciplinary working group will current a brand new, free visualization instrument designed to deal with this drawback. The Hepatic Explorer is an interactive open-source web-based information science utility for monitoring liver toxicity that enables a researcher to each visualize information as an entire and likewise discover red-flag areas. This instrument may very well be of explicit assist through the present pandemic, Buchanan says, as a result of COVID-19 sufferers typically present irregular liver readings that have to be distinguished from the research drug’s results.

The presentation, “Improved Signal Detection and Evaluation Using New Open-Source Interactive Safety Graphics,” will happen Wednesday, August 5, 2020 : 10:00 AM to 11:50 AM EDT. Abstract: https://ww2.amstat.org/meetings/jsm/2020/onlineprogram/AbstractDetails.cfm?abstractid=309756

12) “Nowcasting” and Forecasting COVID-19 **

Modeling the unfold of COVID-19 sometimes takes one in all two approaches: mathematical fashions resembling SIR/SEIR that target the theoretical mechanisms driving the unfold, and statistical fashions extra pushed by information really being noticed. Lily Wang from Iowa State University and collaborators have developed a modeling method that mixes some great benefits of mathematical and statistical fashions to conduct short-term and long-term forecasts. Their “spatio-temporal epidemic model” (STEM) additionally permits researchers to keep in mind the actual traits of every county that have an effect on each illness unfold and fatalities, such because the mobility, age distributions, well being infrastructure, and racial and ethnic demographics. In this presentation, Wang will show the net dashboard they developed based mostly on STEM, which permits customers to visualise, observe, and predict COVID-19 infections and deaths. Wang may even reveal the mannequin’s newest projections for August by December, and present how they evaluate with the CDC’s reported projections.

The presentation, “Spatiotemporal Dynamics, Nowcasting and Forecasting COVID-19 in the United States,” will happen Wednesday, August 5, 2020 : 10:00 AM to 11:50 AM EDT. Abstract: https://ww2.amstat.org/meetings/jsm/2020/onlineprogram/AbstractDetails.cfm?abstractid=309454

13) Text-mining information articles to foretell inventory returns

Advances in information science and machine studying have made it potential to mine monumental units of stories articles and different textual content information to seize subtleties and sentiments of the language inside. This energy of this technique, nevertheless, has but to be absolutely utilized to the sector of finance. Zheng Tracy Ke from Harvard University, Dacheng Xiu from University of Chicago, and their colleagues have developed a pure language processing technique that is particularly designed to mine textual content paperwork to foretell inventory returns. In this speak, the researchers will current outcomes from making use of the brand new methodology to six.7 million articles from the Dow Jones Newswires, some of the actively monitored monetary information streams. They will present that their method can be utilized to analyze how inventory costs reply to the information, and likewise has worth for sensible asset administration.

The presentation, “Predicting Returns with Text Data,” will happen Wednesday, August 5, 2020 : 10:00 AM to 11:50 AM EDT. Abstract: https://ww2.amstat.org/meetings/jsm/2020/onlineprogram/AbstractDetails.cfm?abstractid=309567

THURSDAY HIGHLIGHTS

14) Group testing for COVID-19: What is it, and the way may it’s applied extra successfully? **

Everyone agrees that COVID-19 testing is the important thing to containing the coronavirus, saving lives, and reopening the economic system. The drawback: There merely aren’t sufficient checks for all of the individuals we wish to display screen. The resolution? “Just group it,” says Chris Bilder of the University of Nebraska-Lincoln. His work with colleagues on “group testing,” also called “specimen pooling” and “pooled testing,” has proven that this intelligent statistical technique is technically possible and will make accessible testing sources go loads additional. In this introductory overview lecture, Bilder will clarify what group testing is, talk about its historical past and challenges, present how it’s getting used, and clarify the way it may very well be applied extra successfully.

The presentation, “JUST GROUP IT. Group Testing for Identification,” will happen Thursday, August 6, 2020 : 10:00 AM to 11:50 AM EDT. Abstract: https://ww2.amstat.org/meetings/jsm/2020/onlineprogram/AbstractDetails.cfm?abstractid=314408

15) Panel on statistical significance and P-values

In this session organized by Deborah Mayo of Virginia Tech, 4 panelists will revisit the talk round statistical significance and P-values, together with the presuppositions of criticisms which were raised, the ramifications of reforms which were proposed, and an appraisal of other strategies.

The session, “P-Values and “Statistical Significance”: Deconstructing the Arguments,” will happen Thursday, August 6, 2020 : 10:00 AM to 11:50 AM EDT. Abstract: https://ww2.amstat.org/meetings/jsm/2020/onlineprogram/AbstractDetails.cfm?abstractid=309634

16) COVID-19 Infectious illness modeling and statistics: myths, maxims and mobilization **

In this 40-minute ASA Public Lecture, famous epidemiologist Britta Jewell from the MRC Centre for Global Disease Analysis at Imperial College London and famous statistician Nick Jewell from the London School of Hygiene & Tropical Medicine will talk about info and myths surrounding the COVID-19 pandemic within the US. They may even share insights that may be gleaned from mathematical fashions and statistical info and can clarify what the nation must do subsequent round gathering and decoding information.

The presentation, “COVID-19: Infectious Disease Modeling and Statistics-Myths, Maxims and Mobilization,” will happen Thursday, August 6, 2020: 12:00 PM – 1:00 PM EDT. Details: https://ww2.amstat.org/meetings/jsm/2020/onlineprogram/ActivityDetails.cfm?SessionID=220178

17) New mannequin for contact tracing and illness unfold **

Traditional infectious illness fashions assume that individuals combine randomly in a inhabitants, and that everybody is equally prone to come into contact everybody else. In actuality, nevertheless, we every have our personal community of contacts that we’re extra prone to combine with, which is why contact tracing is so vital throughout pandemics. Fan Bu at Duke University and colleagues got down to develop a brand new technique combining each approaches and likewise bettering on previous fashions. Their technique can account for a way our networks evolve throughout an epidemic as our habits modifications, and the way this in flip impacts illness unfold. The mannequin can even deal with real-world conditions with solely partial information and the place uncertainty is vital. In this presentation, Bu will use actual information from a 2013 flu transmission to point out how the brand new technique can incorporate high-tech contact tracing information to enhance modeling and forecasting.

The presentation, “Likelihood-Based Inference for Partially Observed Epidemics on Dynamic Networks,” will happen Thursday, August 6, 2020 : 1:00 PM to 2:50 PM EDT. Abstract: https://ww2.amstat.org/meetings/jsm/2020/onlineprogram/AbstractDetails.cfm?abstractid=309898

18) Who are the scientific grant gatekeepers?

The progress of scientific analysis depends closely on grant peer evaluate – a course of during which scientists comply with spend important quantities of time anonymously evaluating grant purposes submitted by different researchers. Despite its significance, there was little or no information on traits of scientists doing the opinions, so a complete survey of scientists was developed and administered to study extra. Stephen Gallo of the American Institute of Biological Sciences will current outcomes displaying an uneven distribution of grant peer evaluate participation, with almost half of all reported opinions carried out by lower than 1 / 4 of all respondents, and most reporting that they had been working at most capability. Implications for the way forward for science will likely be mentioned.

The presentation, “The Participation and Motivations of Grant Peer Reviewers,” will happen Thursday, August 6, 2020 : 3:00 PM to 4:50 PM EDT. Abstract: https://ww2.amstat.org/meetings/jsm/2020/onlineprogram/AbstractDetails.cfm?abstractid=309505

19) Statistical fashions for evaluating state opioid insurance policies

To deal with the U.S. opioid disaster, public well being consultants in several states are attempting out a variety of coverage approaches, resembling educating physicians, monitoring prescriptions, and making opioid overdose remedy accessible. Published research that consider these totally different approaches have elevated greater than ten-fold prior to now 15 years, nevertheless it’s nonetheless troublesome to statistically evaluate the success of various packages throughout states. In this presentation, Beth Ann Griffin from the RAND Corporation will talk about work with colleagues that examined these statistical strategies, particularly when a state’s explicit opioid issues affect the insurance policies it chooses. Simulation outcomes from real-world information will likely be offered.

The presentation, “Evaluating Methods to Estimate the Effect of State Laws on Opioid-Related Outcomes in the Presence of Selection Bias,” will happen Thursday, August 6, 2020 : 3:00 PM to 4:50 PM EDT. Abstract: https://ww2.amstat.org/meetings/jsm/2020/onlineprogram/AbstractDetails.cfm?abstractid=309239

###

MORE MEETING INFORMATION

USEFUL LINKS

Main assembly web page: https://ww2.amstat.org/meetings/jsm/2020/

Online program web page: https://ww2.amstat.org/meetings/jsm/2020/onlineprogram/index.cfm

PRESS REGISTRATION

The American Statistical Association will present free registration to all employees journalists, freelance journalists, and pupil journalists. Press credentials are authorized on the discretion of the ASA. For entry, please e mail publicrelations@amstat.org.

ABOUT JSM 2020

Occurring yearly since 1974, JSM is a joint effort of the American Statistical Association, International Biometric Society (ENAR and WNAR), Institute of Mathematical Statistics, Statistical Society of Canada, International Chinese Statistical Association, International Indian Statistical Association, Korean International Statistical Society, International Society for Bayesian Analysis, Royal Statistical Society, and International Statistical Institute. JSM actions embody oral shows, panel periods, poster shows, skilled growth programs, an exhibit corridor, a profession service, society and part enterprise conferences, committee conferences, social actions and networking alternatives.

ABOUT ASA

The American Statistical Association is the world’s largest group of statisticians and the oldest constantly working skilled science society within the United States. Its members serve in business, authorities and academia in additional than 90 international locations, advancing analysis and selling sound statistical observe to tell public coverage and enhance human welfare. For extra info, please go to the ASA web site at http://www.amstat.org.

FOR MORE INFORMATION

Regina Nuzzo, Senior Advisor for Statistics Communication

regina@amstat.org

Steve Pierson, Director of Science Policy

(703) 302-1841

LEAVE A REPLY

Please enter your comment!
Please enter your name here