Statistical inference with missing values + ggplot2 workshops
18:15 - 19:45 - ggplot2 workshop
20:00 - 20:30 - Statistical inference with missing values, Wei Jiang
20:30 - ... - pizza
Spotkania Entuzjastów R are back with a new meetup formula!
The December meeting will be divided into two parts, a ggplot2 workshop and Wei Jiang's presentation entitled "Statistical inference with missing values".
The first part is a hands-on workshop (in Polish) to make sure that participants (especially beginners) have the opportunity to learn something new and try it out in practice.
Afterward, will be a more advanced presentation of the invited guest, who will demonstrate interesting applications of R. However, it is worth to come regardless of your R level, surely you will find something for yourself!
The last part of the meeting will be the networking session with pizza, there will be a chance to get to know each other better and exchange experiences.
ggplot2 workshops registration (in Polish):
Workshops are open to everyone, but we would like to ask you to fill in the form.
This will help us to adjust the advancement level of the workshops and divide participants into groups with a similar level of R knowledge. In addition, we will be able to send further details, such as room numbers or links to materials to your e-mail.
# Statistical inference with missing values
Missing data exist in almost all areas of empirical research. There are various reasons why missing data may occur, including survey non-response, unavailability of measurements, and lost data. In this presentation, I will first share my experience on how to do parametric estimation with missing covariates, based on likelihood methods and Expectation-Maximization algorithm. Then I will focus on recent results of variable selection in high-dimensional settings with missing observations among the covariates. To address this relatively understudied problem, we propose a new synergistic procedure – adaptive Bayesian SLOPE – which effectively combines the SLOPE method (sorted l1 regularization) together with the Spike-and-Slab LASSO method. Through extensive simulations, we demonstrate satisfactory performance in terms of power, FDR and estimation bias under a wide range of scenarios. Finally, we analyze a real dataset consisting of patients from Paris hospitals who underwent a severe trauma, where we show excellent performance in predicting platelet levels.
Wei Jiang is currently pursuing PhD degree in Statistics at Ecole Polytechnique, France. Her research mainly focuses on statistical inference with missing values and applications on medical dataset. Fan of R, eager to learn more about data science and mathematics.