A 3-Day Remote Workshop Taught by Towhid Islam, PhD., University Research Leadership Chair and Professor, Department of Marketing and Consumer Studies, University of Guelph, Canada.
see the link for details: https://dataorbit.ca/course/12
Workshop: Thursday, July 14, 2022 – Saturday, July 16, 2022.
Live Lecture and Lab Session, 10 am–2:30 pm EST (New York Time) each day with 30 minutes break at noon. Video-recorded versions will be made available after each session.
Payment: The fee of Canadian $700 Regular, $500 for Students, includes lecture material, R codes, and data. PayPal, major credit cards, and e-transfer are accepted.
Description
This workshop will provide the foundations of machine learning procedures (MLPs) that are used mainly for predictions and cover recent developments of MLPs that can be used for causal inference and hypotheses testing from observational data. The technological revolution of the internet has generated a repository of digital observational data as a byproduct, and this revolution has been complemented by an influx of MLPs from computer science, leading to an emerging new field of computational social science. In the social sciences, empirical analyses typically estimate the causal effect of implementing a policy, changing a price, running an advertisement, or introducing new products. The objective here is the integration of prediction and explanation into a data driven computational social science. Researchers will learn about the process and implicit flow of action behind the MLPs; the ability to build models for predictions and causal effects for policy decisions and to explore and visualize novel and robust non-linear and interdependent patterns that may aid in theory testing and theory building by blending data and theory. The workshop will use R, a free and open-source package for statistical analysis with excellent graphics capabilities.
------------------------------
Towhidul Islam
University Research Leadership Chair & Professor
University of Guelph
Guelph ON
------------------------------