Warwick Mathematics Institute – Lockdown Masterclasses on Mathematical Thinking

Prof Minhyong Kim, WMI: "I am writing today to call your attention to a small initiative by the Mathematics Institute to aid the members of the local community in meeting the educational needs of young people during this difficult period...this webpage for the Warwick Mathematics Institute Lockdown Masterclasses on Mathematical Thinking (for children, teens, and their adults). I do hope many of you will participate with your family members, and will end up finding the sessions useful and enjoyable."

The first session is on 4^th June at 4pm, please register at this link.

More resources on Prof Minhyong Kim's Warwick outreach page.

Statistics in Sports:

• Premier League Tables: What do they reflect? Could there be an alternative way of calculating such tables reflecting the performance of the teams better? Check out Prof David Firth' work on the alt-3 - the real league table to find out more.
• Keeping track: Are you interested in open source software for the analysis of running, cycling and swimming data from GPS-enabled tracking devices? Dr Ioannis Kosmidis has built this for you in trackeRapp; an R shiny web-interface. For a local installation of the web-interface you can install the trackeRapp R package, and for more advanced statistical modelling of such data you can directly use the core package trackeR. An introduction to trackeRapp and trackeR can be watched in Ioannis' Pint of Science 2019 talk. If you are not prepared to move around so much yourself, see also the data analysis projects in Package 2 with suggestions where to find tracking data from animals!
• Simulating football (soccer) matches: Can you build a real-time match simulator based on sequences of touch-ball events? Santhosh Naranayan wrote his PhD thesis about this and related topics and presented some of his innovative work at this year's OptaPro Analytics Forum.
• The sport section: The magazine Significance, which publishes articles on topics of statistical interest presented at a level suited for a general audience, has a Sport Section.
• Other side of the pond: The American statistician Nate Siver's career began with forecasting the performance of Major League Baseball players. While he and his team are now more busy with quantitative political analyses, there is the sport section of fivethirtyeight.

Music Data Analysis:

• Music recommendations: Have you ever wondered how to evaluate music recommendations? A starting point can be the Million Song Dataset is a freely-available collection of audio features and metadata for a million contemporary popular music tracks.
• To skip or not to skip: Can you predict what users are doing? The WSDM Cup music streaming session data set allows you to try your luck.
• Music data analysis in R: This short course by Bruna Wundervald and Julio Trecenti provides some examples and sample code of exploratory and model bases analysis of Spotify data.
• Polyphonic recordings: If you do not know what that is it is time to find out. The OpenMIC data set is for multi-instrument recognition in polyphonic recordings, a fundamental problem in music information retrieval.
• Visual music project: Ned Resnikoff uses signal processing methods to understand and visualise music. Check out the beautiful pictures using a wavelet analysis approach in the spectogram project and watch music visualisations.
• Data science for music: The TidyTuesday Project has had a few music-themed weeks including about a BBC survey by way of Simon Jockers at Datawrapper on Rap Artists, about Christmas Music Billboards and about Spotify Songs.
• Analysis and visualisation of musical lyrics: An algorithm to find out how Hamilton rhymes to blur musical lines and and interactive visualisation of every line in Hamilton.

Statistics of Medical Testing and Implications:

Medical testing in a complex process. What exactly are you testing for and what does it have to do with the disease? (Often, tasing is about markers linked to the disease.) How reliable is the test? This includes finding the disease when present (sensitivity) and not indicating disease when it is not present (specificity) all expressed in probabilities. Was is the consequences of an incorrect test result? Are test result always as useful as one might think even if correct? (For example, do you want to know at age 18 that you may, with a probability of 63%, develop Parkinson at age 53+/-5 years?) Under which conditions are they useful? These were just some of the most important questions. There is a lot to consider and discuss when it comes to medical testing.

There is a lot of debate at the moment in conjunction with Covid-19 disease testing, both to determine if someone is currently infected (swab test for the virus) and to determine if someone has had the disease previously (blood test for antibodies). Here are some links that focus on the statistical and societal aspects of these tests and their use.

• Comparing and assessing Covid-19 tests
• Test reliability & Bayes theorem
• Immunity certificates (virtual event on 3^rd June), Test, trace and trust: digital technologies and Covid-19 response  (recording) and Contact tracing: to centralise or not (recording), all from the Ada Lovelace Institute, an independent research and deliberative body with a mission to ensure data and AI work for people and society

More on Covid-19 will be posted here soon:

Please stay tuned. In the meantime you can find out more about Covid-19 research and activities across the University of Warwick.