[Pols-l] CRMDA workgroups Spring 2019, beginning this week!

Paul Johnson pauljohn at ku.edu
Tue Jan 22 10:25:40 CST 2019


Dear Friends

The CRMDA is offering methods workgroups for both Python and Big Data 
research.  Ben Sherwood is the Faculty leader for the Big Data Research 
workgroup. Everybody to know that on Friday, January 25, 2019, at 2PM, 
the first meeting of the Big Data group will be held. The bi-weekly 
meetings of the Python workgroup will begin on February 1, 2019.

I plan to be a participant/student in both.  We hope many new users will 
join and contribute. Signup and schedule information is available online 
https://crmda.ku.edu/workgroups.

Here is the Big Data group message from Professor Sherwood:
###
We are continuing the big data seminar this Friday at 2pm and we will be 
meeting every other Friday. This Friday we will do a review of what was 
discussed last semester and an overview of topics for this semester.

Last semester, we discussed likelihood ratio tests, AIC, BIC and ridge 
regression. This semester we will focus much more on Lasso and some of 
the Lasso extensions such as group Lasso.

I will do a review of the material we discussed last semester and give 
an overview of topics I wish to discuss this semester. This will include

1. Lasso
2. Group Lasso
3. Using Lasso with missing data.
4. Inference (p-values, standard errors etc.) with Lasso and why it is hard.
5. Non-convex penalty functions such as SCAD.

This first meeting would also be a good time to discuss other related 
issues that people would like to review.

Ben Sherwood
###

The Python Text Analysis workgroup, in which the leader is Jon Lamb of 
the KU English Department, will begin on February 1, in Watson 455, at 1PM.

Here is the Python group message from Professor Lamb:
###
This workgroup is designed for anyone who wants to learn digital text 
analysis and integrate it into their research. We will explore advanced 
techniques, showing examples of methods prominent in the quantitative 
study of texts: Topic Modeling, Random Forests, Principal Components 
Analysis, and more.

Everyone is welcome in this group; no prior programming experience is 
required. The group will definitely be a workgroup, where we will 
discuss methods, practical questions and intellectual problems that 
arise from them, and how to implement them into text analysis research.

Jon Lamb
###

pj

-- 
Paul E. Johnson			University of Kansas	     	
Professor			Director, Center for Research
Political Science		Methods & Data Analysis (CRMDA)
http://pj.freefaculty.org	http://crmda.ku.edu
email: pauljohn at ku.edu
Address: CRMDA
	 Watson Library, Suite 470
	 1425 Jayhawk Blvd.	
    	 Lawrence, Kansas
  	 66045-7594
   	 Ph: (785) 864-8215



More information about the Pols-l mailing list