POPULARITY
Statistician David Spiegelhalter is no stranger to AI – he used it to help him research his recent book and, back in the late 70s, he helped develop foundational algorithms for the tech. So, he understands the pandora’s box that technology can represent, as well as the uncertainty embedded in its future development. Spiegelhalter sits down with Oz to unpack how we should interpret AI predictions, why better data matters and why we should consciously embrace uncertainty in our own lives.See omnystudio.com/listener for privacy information.
David Spiegelhalter is one of the world's most important figures in statistics. He's an emeritus professor of statistics in the Centre for Mathematical Studies at the University of Cambridge and he's the author of The Art of Uncertainty: How to Navigate Chance, Ignorance, Risk and Luck. Spiegelhalter is committed to making mathematics more accessible, and he joins My Wildest Prediction to talk about probabilities, how to deal with uncertainty and artificial intelligence. My Wildest Prediction is a podcast series from Euronews Business where we dare to imagine the future with business and tech visionaries. Hosted on Acast. See acast.com/privacy for more information.
Facebook: www.facebook.com/petofiradio/?ref=pages_you_manage Instagram: www.instagram.com/petofiradio/
Today on We Are Not Amused, Tressa and Taylor once again dive into the topic of true crime. Join us as we discuss Joseph Kappen and Harold Shipman, how science played a part in their murders, and, surprisingly, a lot of statistics. Grab a cup of tea and get ready to be transported back to AP Stats as we discuss these two serial killers.
Link to bioRxiv paper: http://biorxiv.org/cgi/content/short/2023.01.11.523642v1?rss=1 Authors: Durik, M., Sampaio Goncalves, D., Spiegelhalter, C., Messaddeq, N., Keyes, B. Abstract: Cellular senescence is a complex cell state with roles in tumor suppression, embryonic development and wound repair. However, when misregulated, senescence contributes to aging and disease. Here we identify that senescent cells generate/break off large membrane-bound fragments of themselves through cell-to-cell adhesion. We designate these as senescent-cell adhesion fragments (SCAFs) which were present in all types of senescent cell examined. We show they contain many organelles from the original cell, but without nuclear material. Quantitative and dynamic profiling shows that SCAFs are large, may persist for a number of days, but rupture and release their contents onto neighboring cells. Protein profiling identifies that SCAFs contain a complex proteome including immune recruitment factors and damage-associated molecular patterns (DAMPs). Functional studies reveal that SCAFs activate signatures related to wound healing and cancer, and promote invasion and migration. Altogether, we uncover an additional cellular feature of senescent cells, by which they deposit intracellular contents on other cells. We speculate this may aid in boosting immune responses, but in chronic situations, may contribute to debris buildup, inflammaging and age-associated changes. Copy rights belong to original authors. Visit the link for more info Podcast created by Paper Player, LLC
Hosted on Acast. See acast.com/privacy for more information.
How can we make sense of what we're told about risk? We're bombarded with messages on subjects ranging from COVID to the economy from people that range from genuine experts to those with no expertise but strong opinions. On this episode, I'm speaking to Professor David Spiegelhalter.David is Chair of the Winton Centre for Risk and Evidence Communication within the Department of Pure Mathematics and Mathematical Statistics at Cambridge University. The Centre is dedicated to improving the way that quantitative evidence is used in society. Listeners in the UK will almost certainly have seen or heard David. Since the start of the pandemic he's been a regular fixture on TV and radio, helping to make sense of the things we're being told about the virus. In a world of self-appointed experts whose only qualification is from the University of YouTube and untrustworthy politicians telling us they're "following the science", he's been a voice of clarity and common sense. In our discussion, we explore what drives David's interest in statistics, why we like to see connections between things that might not actually be there, why the mantra of “following the science” is nonsensical and whether there is such a thing as coincidence. David also provides plenty of practical tips for communicating and interpreting messages about risk. As you might expect for someone who specialises in risk communication, David is really good at getting his message across in ways we can all understand. My huge thanks to long-time friend of the show Roger Miles, who helped to make this conversation possible.To find out more about David, visit his academic website: https://wintoncentre.maths.cam.ac.uk/about/people/professor-sir-david-spiegelhalter/or his personal website: https://www.statslab.cam.ac.uk/~david/You'll find his books in all good bookstores. For more information, visit:The Art of Statistics — https://www.penguin.co.uk/books/294857/the-art-of-statistics-by-spiegelhalter-david/9780241258767COVID by Numbers — https://www.penguin.co.uk/authors/126755/david-spiegelhalterFor video content, I recommend:Communicating statistics in the time of COVID — https://www.youtube.com/watch?v=JW9plVfanjoFalse Positives — https://www.youtube.com/watch?v=XmiEzi54lBIBe Prepared To Show Your Working — https://www.youtube.com/watch?v=E12_F4xeOHwIn our discussion, we also refer to the episode featuring Tim Harford on using Data to Make Smarter Decisions. You can hear that here: https://www.humanriskpodcast.com/tim-harford-on-using-data/
As a podcaster, I discovered that there are guests for which the hardest is to know when to stop the conversation. They could talk for hours and that would make for at least 10 fantastic episodes. Frank Harrell is one of those guests. To me, our conversation was both fascinating — thanks to Frank's expertise and the width and depth of topics we touched on — and frustrating — I still had a gazillion questions for him! But rest assured, we talked about intent to treat and randomization, proportional odds, clinical trial design, bio stats and covid19, and even which mistakes you should do to learn Bayes stats — yes, you heard right, which mistakes. Anyway, I can't tell you everything here — you'll just have to listen to the episode! A long time Bayesian, Frank is a Professor of Biostatistics in the School of Medicine at Vanderbilt University. His numerous research interests include predictive models and model validation, Bayesian clinical trial design and Bayesian models, drug development, and clinical research. He holds a PhD in biostatistics from the University of North Carolina, and did his Bachelor in mathematics at the University of Alabama in Birmingham. Our theme music is « Good Bayesian », by Baba Brinkman (feat MC Lars and Mega Ran). Check out his awesome work at https://bababrinkman.com/ (https://bababrinkman.com/) ! Thank you to my Patrons for making this episode possible! Yusuke Saito, Avi Bryant, Ero Carrera, Brian Huey, Giuliano Cruz, Tim Gasser, James Wade, Tradd Salvo, Adam Bartonicek, William Benton, Alan O'Donnell, Mark Ormsby, Demetri Pananos, James Ahloy, Jon Berezowski, Robin Taylor, Thomas Wiecki, Chad Scherrer, Nathaniel Neitzke, Zwelithini Tunyiswa, Elea McDonnell Feit, Bertrand Wilden, James Thompson, Stephen Oates, Gian Luca Di Tanna, Jack Wells, Matthew Maldonado, Ian Costley, Ally Salim, Larry Gill, Joshua Duncan, Ian Moran, Paul Oreto, Colin Caprani, George Ho, Colin Carroll, Nathaniel Burbank, Michael Osthege, Rémi Louf, Clive Edelsten, Henri Wallen, Hugo Botha, Vinh Nguyen, Raul Maldonado, Marcin Elantkowski, Tim Radtke, Adam C. Smith, Will Kurt, Andrew Moskowitz, Hector Munoz, Marco Gorelli, Simon Kessell, Bradley Rode, Patrick Kelley, Rick Anderson, Casper de Bruin and Philippe Labonde. Visit https://www.patreon.com/learnbayesstats (https://www.patreon.com/learnbayesstats) to unlock exclusive Bayesian swag ;) Links from the show: Frank's website and courses: https://hbiostat.org/ (https://hbiostat.org/) Frank's blog: https://www.fharrell.com/ (https://www.fharrell.com/) Frank on Twitter: https://twitter.com/f2harrell (https://twitter.com/f2harrell) COVID-19 Randomized Clinical Trial Design: https://hbiostat.org/proj/covid19/ (https://hbiostat.org/proj/covid19/) Frank on GitHub: https://github.com/harrelfe (https://github.com/harrelfe) Regression Modeling Strategies repository: https://github.com/harrelfe/rms (https://github.com/harrelfe/rms) Biostatistics for Biomedical Research repository: https://github.com/harrelfe/bbr (https://github.com/harrelfe/bbr) Bayesian Approaches to Randomized Trials, Spiegelhalter et al.: http://hbiostat.org/papers/Bayes/spi94bay.pdf (http://hbiostat.org/papers/Bayes/spi94bay.pdf) Statistical Rethinking, Richard McElreath: http://xcelab.net/rm/statistical-rethinking/ (http://xcelab.net/rm/statistical-rethinking/) LBS #20, Regression and Other Stories, with Andrew Gelman, Jennifer Hill & Aki Vehtari: https://www.learnbayesstats.com/episode/20-regression-and-other-stories-with-andrew-gelman-jennifer-hill-aki-vehtari (https://www.learnbayesstats.com/episode/20-regression-and-other-stories-with-andrew-gelman-jennifer-hill-aki-vehtari) David Spiegelhalter, The Art of Statistics -- Learning from Data: https://www.amazon.fr/Art-Statistics-Learning-Data/dp/0241398630 (https://www.amazon.fr/Art-Statistics-Learning-Data/dp/0241398630) This podcast uses the following third-party services for analysis: Podcorn -... Support this podcast
Steve Spiegelhalter is the North American Investigations Practice Leader and Managing Director at Alvarez & Marsal. As a former federal prosecutor with the US Department of Justice's Criminal Division, Fraud Section, and the Foreign Corrupt Practices Act (FCPA) Unit, Steve has intimate experience in investigating complex criminal and civil affairs and implementing compliance programs. He joins Vince Walden to discuss the future of conducting internal investigations. Steve talks about the improvements GCs and CCOs have made in internal investigations over the last five years. They have evolved their in-house skills and resources. Additionally, they have gotten better at interacting with external counsel to solve matters more efficiently. COVID-19 has highlighted that foreign corrupt practices and corruption are long-term risks that are becoming more prominent. There has been a rise in fraud issues since the workplace has shifted to remote. Behaviors of malpractice that would have gone unnoticed are now being laid bare. Resources Steve Spiegelhalter on LinkedIn AlvarezandMarsal.com
Professor Sir David Spiegelhalter is Chair of the Winton Centre for Risk and Evidence Communication in the University of Cambridge, which aims to improve the way that statistical evidence is used by health professionals, patients, lawyers and judges, media and policy-makers. He advises organisations and government agencies on risk communication and is a regular media commentator on statistical issues, with a particular focus on communicating uncertainty.He has over 200 refereed publications and is co-author of 6 textbooks, as well as The Norm Chronicles (with Michael Blastland), Sex by Numbers, and The Art of Statistics. He works extensively with the media, and presented the BBC4 documentaries ‘Tails you Win: the Science of Chance” and the award-winning “Climate Change by Numbers”. He was elected Fellow of the Royal Society in 2005, knighted in 2014 for services to medical statistics, and was President of the Royal Statistical Society for 2017-2018.
Professor Sir David Spiegelhalter is Chair of the Winton Centre for Risk and Evidence Communication in the University of Cambridge, which aims to improve the way that statistical evidence is used by health professionals, patients, lawyers and judges, media and policy-makers. He advises organisations and government agencies on risk communication and is a regular media commentator on statistical issues, with a particular focus on communicating uncertainty.He has over 200 refereed publications and is co-author of 6 textbooks, as well as The Norm Chronicles (with Michael Blastland), Sex by Numbers, and The Art of Statistics. He works extensively with the media, and presented the BBC4 documentaries ‘Tails you Win: the Science of Chance” and the award-winning “Climate Change by Numbers”. He was elected Fellow of the Royal Society in 2005, knighted in 2014 for services to medical statistics, and was President of the Royal Statistical Society for 2017-2018.
This Week in Machine Learning & Artificial Intelligence (AI) Podcast
In this, the second episode of our NeurIPS series, we’re joined by David Spiegelhalter, Chair of Winton Center for Risk and Evidence Communication at Cambridge University and President of the Royal Statistical Society. David, an invited speaker at NeurIPS, presented on “Making Algorithms Trustworthy: What Can Statistical Science Contribute to Transparency, Explanation and Validation?”. In our conversation, we explore the nuanced difference between being trusted and being trustworthy, and its implications for those building AI systems. We also dig into how we can evaluate trustworthiness, which David breaks into four phases, the inspiration for which he drew from British philosopher Onora O'Neill's ideas around 'intelligent transparency’. The complete show notes for this episode can be found at twimlai.com/talk/212. For more information on the NeurIPS series, visit twimlai.com/neurips2018.
Every day we are bombarded with statistics about sex. How many times we think of it a day, how many times we do it, and with how many people. But how do we know which of those numbers can be believed? David Spiegelhalter, one of our favourite statisticians, has written a book all about the stats of sex, called "Sex by numbers". In this podcast Spiegelhalter gives us some of his favourite snippets from the book, which are as informative as they are entertaining. You can also watch our interview as a video or read the associated article – https://plus.maths.org/content/sexual-statistics
Every day we are bombarded with statistics about sex. How many times we think of it a day, how many times we do it, and with how many people. But how do we know which of those numbers can be believed? David Spiegelhalter, one of our favourite statisticians, has written a book all about the stats of sex, called "Sex by numbers". In this podcast Spiegelhalter gives us some of his favourite snippets from the book, which are as informative as they are entertaining. You can also watch our interview as a video or read the associated article – https://plus.maths.org/content/sexual-statistics
While they aren’t as unpopular as politicians or journalists, people who work with statistics come in for their share of abuse. “Figures lie and liars figure,” goes one maxim. And don’t forget, “There are three kinds of lies: lies, damned lies, and statistics." But some people are the good guys, doing their best to combat the flawed or dishonest use of numbers. One of those good guys is David Spiegelhalter, professor of the public understanding of risk in the Statistical Laboratory in the Centre for Mathematical Sciences at the University of Cambridge and current president of the Royal Statistical Society. Spiegelhalter, the subject of this Social Science Bites podcast, even cops to being a bit of an “evidence policeman” because on occasion even he spends some of his time “going around telling people off for bad behavior.” There is bad behavior to police. “There’s always been the use of statistics and numbers and facts as rhetorical devices to try and get people’s opinion across, and to in a sense manipulate our emotions and feelings on things,” he tells interviewer David Edmonds. “People might still think that statistics and numbers are cold, hard facts but they’re soft, fluffy things. They can be manipulated and changed, made to look big, made to look small, all depending on the story that someone wants to tell.” Asked at one point if he even accepts that there are ‘facts,’ Spiegelhalter gives a nuanced yes. “I’m not going to get into the whole discussion about ‘what is truth,’ although it’s amazing how quick you do go down that line. No, there are facts, and I really value them.” Despite that policing role, Spiegelhalter explain, his methods are less prescriptive and more educational, working to get others to ask key questions such as “What am I not being told?” and “Why I am hearing this?” The goal is less to track down every bit of fake news in the world, and more to inoculate others against its influence. One part of that, for example, is determining what communicators and organizations to trust. Spiegelhalter, acknowledging his debt to Onora O'Neill, an emeritus professor of philosophy at the University of Cambridge, argues that organizations themselves shouldn’t strive to be trusted, but to show trustworthy attributes. This goes beyond things like “fishbowl transparency,” where you lard your website with every imaginable factoid, but actively making sure people can get to your information, understand it and they can assess how reliable it is. That ‘understanding’ part of the process is what Spigelhalter pursues as part of chairing the Winton Centre for Risk and Evidence Communication, which is dedicated to improving the way that quantitative evidence is used in society. In that role he’s become a public face of honest use of numbers, as evidenced by his role as presenter of the BBC4 documentaries Tails you Win: the Science of Chance and Climate Change by Numbers. His own research focuses on health-related use of statistics and statistical methods, and while that might include Bayesian inference using Gibbs samplinig, it can also encompass the focus of his 2015 book, Sex by Numbers.
Rhianna Dhillon brings you another seriously interesting story from Radio 4. This week, luck. Whether we believe in luck or not, we do use the word- a lot! More as a figure of speech than an article of faith perhaps but some do pray for luck, others fantasise about it - and bad luck or misfortune is a staple of comedy Can luck be said to exist as some force in our lives and if so, what is its nature? How have people thought about luck in the past and what's changed today? Can you bring good luck upon yourself - there's a school of thought these days that thinks you can without appealing to the divine or supernatural. In Good Luck Professor Spiegelhalter, the Winton Professor for the Public Understanding of Risk at Cambridge University looks at notions of luck in gambling, traces the origins of how we think about fate and fortune, the religious and psychological view of luck and how the emergence of theory of probability changed our view of it. He is convinced by the philosopher Angie Hobbs that there is one form of luck it is rational to believe in and by psychologist Richard Wiseman that there is a secular solution to bringing about good fortune in your life. Good Luck Professor Spiegelhalter, is presented by David Spiegelhalter and produced in Salford by Kevin Mousley.
Award-winning broadcaster Trevor Dann presents a new series for Cambridge 105 in which he interviews some of the university city’s brightest stars. The first edition features statistician Sir David John Spiegelhalter, OBE FRS and Master of Selwyn College Roger Mosey.
David Spiegelhalter's proper title is Professor of the Public Understanding of Risk. He is in two minds (literally) about playing it safe or chucking caution to the wind. Decisions, decisions!? Are bacon sandwiches really that dangerous and is it wise to drive when you love cycling? David shows us how to use statistics to face up to life's major risks.
Mathematik, Informatik und Statistik - Open Access LMU - Teil 02/03
In this paper we consider regression models for count data allowing for overdispersion in a Bayesian framework. We account for unobserved heterogeneity in the data in two ways. On the one hand, we consider more flexible models than a common Poisson model allowing for overdispersion in different ways. In particular, the negative binomial and the generalized Poisson distribution are addressed where overdispersion is modelled by an additional model parameter. Further, zero-inflated models in which overdispersion is assumed to be caused by an excessive number of zeros are discussed. On the other hand, extra spatial variability in the data is taken into account by adding spatial random effects to the models. This approach allows for an underlying spatial dependency structure which is modelled using a conditional autoregressive prior based on Pettitt et al. (2002). In an application the presented models are used to analyse the number of invasive meningococcal disease cases in Germany in the year 2004. Models are compared according to the deviance information criterion (DIC) suggested by Spiegelhalter et al. (2002) and using proper scoring rules, see for example Gneiting and Raftery (2004). We observe a rather high degree of overdispersion in the data which is captured best by the GP model when spatial effects are neglected. While the addition of spatial effects to the models allowing for overdispersion gives no or only little improvement, a spatial Poisson model is to be preferred over all other models according to the considered criteria.
Mathematik, Informatik und Statistik - Open Access LMU - Teil 02/03
In this paper models for claim frequency and claim size in non-life insurance are considered. Both covariates and spatial random e ects are included allowing the modelling of a spatial dependency pattern. We assume a Poisson model for the number of claims, while claim size is modelled using a Gamma distribution. However, in contrast to the usual compound Poisson model going back to Lundberg (1903), we allow for dependencies between claim size and claim frequency. Both models for the individual and average claim sizes of a policyholder are considered. A fully Bayesian approach is followed, parameters are estimated using Markov Chain Monte Carlo (MCMC). The issue of model comparison is thoroughly addressed. Besides the deviance information criterion suggested by Spiegelhalter et al. (2002), the predictive model choice criterion (Gelfand and Ghosh (1998)) and proper scoring rules (Gneiting and Raftery (2005)) based on the posterior predictive distribution are investigated. We give an application to a comprehensive data set from a German car insurance company. The inclusion of spatial e ects significantly improves the models for both claim frequency and claim size and also leads to more accurate predictions of the total claim sizes. Further we quantify the significant number of claims e ects on claim size.
Mathematik, Informatik und Statistik - Open Access LMU - Teil 02/03
In this paper we consider regression models for count data allowing for overdispersion in a Bayesian framework. Besides the inclusion of covariates, spatial effects are incorporated and modelled using a proper Gaussian conditional autoregressive prior based on Pettitt et al. (2002). Apart from the Poisson regression model, the negative binomial and the generalized Poisson regression model are addressed. Further, zero-inflated models combined with the Poisson and generalized Poisson distribution are discussed.In an application to a data set from a German car insurance company we use the presented models to analyse the expected number of claims. Models are compared according to the deviance information criterion (DIC) suggested by Spiegelhalter et al. (2002). To assess the model fit we use posterior predictive p-values proposed by Gelman et al. (1996). For this data set no significant spatial effects are observed, however the models allowing for overdispersion perform better than a simple Poisson regression model.
Mathematik, Informatik und Statistik - Open Access LMU - Teil 02/03
In this paper we model absolute price changes of an option on the XETRA DAX index based on quote-by-quote data from the EUREX exchange. In contrast to other authors, we focus on a parameter-driven model for this purpose and use a Poisson Generalized Linear Model (GLM) with a latent AR(1) process in the mean, which accounts for autocorrelation and overdispersion in the data. Parameter estimation is carried out by Markov Chain Monte Carlo methods using the WinBUGS software. In a Bayesian context, we prove the superiority of this modelling approach compared to an ordinary Poisson-GLM and to a complex Poisson-GLM with heterogeneous variance structure (but without taking into account any autocorrelations) by using the deviance information criterion (DIC) as proposed by Spiegelhalter et al. (2002). We include a broad range of explanatory variables into our regression modelling for which we also consider interaction effects: While, according to our modelling results, the price development of the underlying, the intrinsic value of the option at the time of the trade, the number of new quotations between two price changes, the time between two price changes and the Bid-Ask spread have significant effects on the size of the price changes, this is not the case for the remaining time to maturity of the option. By giving possible interpretations of our modelling results we also provide an empirical contribution to the understanding of the microstructure of option markets.
Mathematik, Informatik und Statistik - Open Access LMU - Teil 02/03
In market microstructure theory the effect of time between consecutive transactions and trade volume on transaction price changes of exchange traded shares and options has been considered (e.g. Diamond and Verecchia (1987) and Easley and O'Hara (1987)). The goal of this paper is to investigate if these theoretical considerations can be supported by a statistical analysis of data on transaction price changes of options on shares of the Bayer AG in 1993-94. For this appropriate regression models with non linear and interaction effects are developed to study the influence of trade volume, time between trades, intrinsic value of an option at trading time and price development of the underlying share on the absolute transation price change of an option. Since price changes are measured in ticks yield count data structure, we use in a first analysis ordinary Poisson generalized linear models (GLM) ignoring the time series structure of the data. In a second analysis these Poisson GLM's are extended to allow for an additional AR(1) latent process in the mean which accounts for the time series structure. Parameter estimation in this extended model is not straight forward and we use Markov Chain Monte Carlo (MCMC) methods. The extended Poisson GLM is compared to the ordinary Poisson GLM in a Bayesian setting using the deviance information criterion (DIC) developed by Spiegelhalter et al. (2002). With regard to market microstructure theory the results of the analysis support the expected effect of time between trades on absolute option price changes but not for trade volume in this data set.