Lectures and seminars Workshop “Introduction to Python for Health Researchers”, arranged by the Doctoral Programme in Epidemiology at Karolinska Institutet
1-Day Workshop: Introduction to Python for Health Researchers
Aim: The aim of the workshop is to provide a basic understanding of how Python language can be used by health researchers.
Information and registration: https://bit.ly/2F9N7BV
Date: December 14, 2020
Time: 9:00-12:00 + 13:00-16:00
Location: Zoom , link: https://ki-se.zoom.us/j/63345893602
Teacher: Nicola Orsini, PhD, Dept. Global Public Health, Karolinska Institutet
Target group: Doctoral students, Master-students, Junior and Senior researchers
Requirements: The course assumes that the participant already had a decent undergraduate or postgraduate level introduction to probability and statistics. The participants is expected to have sufficient familiarity with the computer to be able to install Python (https://www.python.org), Jupyter Notebook (https://jupyter.org/) as well as the following modules: pandas, matplotlib, numpy, scipy, statsmodels (https://pypi.org/project/pip/).
After successfully completing this workshop participants will be able to:
1. import and describe the data
2. create a table of summary statistics
3. conduct statistical tests (t-test, Chi-Square test)
4. produce high quality figures of statistics
5. fit regression models (linear, logistic)
6. modelling quantitative predictors using splines
7. conduct statistical inference based on the fitted model
8. control high quality graphs of statistical inference based on linear combinations
9. simulate plausible data generating mechanisms
10. write new functions, looping, comprehensions
The course is a full-time hands-on practice of Python language answering relevant health related questions based on either empirical or simulated data. The participant will learn how to import a dataset, create visualizations of distributions and statistics, estimation using popular regression models (linear, logistic), inference (likelihood based statistical tests, pointwise confidence intervals) on predicted responses or changes in predicted responses, draw pseudo-random values from theoretical probability distributions, Monte-Carlo simulations of common data generating mechanisms (interaction, non-linearity), and basic elements of programming such as creating new functions and avoid looping using comprehensions.
For questions please email Nicola Orsini (firstname.lastname@example.org), Biostatistics Team, Department of Global Public Health , Karolinska Institutet (http://ki.se/en/phs/biostatistics-team).