I just returned from PSI conference 2018. I got a chance to give an oral present at session “Step into the Real World”. I shared my experience from taking part in Roche Advanced Analytics Data (RAAD) Challenge. The take-home message is “with available wealthy real-world data and the power of machine learning for such data, statisticians would be benefit to know machine learning methods”. You can read my slides “Predict survival for cancer patients using real world data: simple model or advanced machine learning?” here. The presentation was recorded, so you can also go to PSI “Video-on-Demand Library” to watch the presentation if you are a PSI member (may need to wait for a couple of weeks for video be uploaded).
Data science and/or machine learning gains more attention at such stats conference, there were at least 6 talks on this topic including one keynote speaker and a dedicated session “Rise of the machines”. People have different perceptive on data science/machine learning. It would be interesting to see if machine learning will be more accepted by statisticians in future.
Estimands, Bayesian and Graphics (visualization) are all hot topics. R shiny has been discussed and presented in different sessions. It seems Pharmaceutical companies all use R shiny, some are in early stage (basic plots), some are more advanced, some even published Shiny app at a public domain (outside the company). I think it would be great if those Shiny apps developed by statisticians (in pharmaceutical companies) can be shared widely, especially for the apps related to methodology. One action for me is to publish my two Shiny apps (DDCP and “Predictive Probability”) at www.shinyapps.io.
Overall, it is a very good conference for sharing, learning, and networking. Next year (2019), the PSI conference will be held in London!