Summer School - Data Analysis and Statistics, 22 May - 9 June University of Kansas, USA

Publish Date: Apr 03, 2017

Deadline: May 15, 2017

Event Dates: from May 22, 2017 12:00 to Jun 09, 2017 12:00

About the summer school

Graduate students, faculty, post doctoral fellows, staff, and others are invited to attend the KU Stats Camp. This year, we are offering three weeks of sessions about R, Stata, and Structural Equation Modeling. KU Stats Camp begins in the last week of May and continues through the first two weeks of June. Participants are encouraged to enroll for all sessions in each week. 

Registration is open to KU students, faculty, and staff, as well as corporate employees or affiliates of other universities.  The daily price of attendance for KU affiliates is $50, while the daily price for others is $70.

Summer school agenda

Day 1—Installation & Getting to Know R

Instructor: Dr. Paul Johnson, CRMDA Director and Professor, Political Science
 Installation
 R documentation, interacting with the R help system, packages
 User interfaces: Comparing Windows R, Emacs, Notepad++, Macintosh
 Line aRt: Illustrating functions, create publication quality line art and graphic displays

Day 2—Establishing a Replicable Research Process

Instructor: Dr. Paul Johnson, CRMDA Director and Professor, Political Science
 Data impact and Recode data: Wrestling with numerical, text, and factor variables
 Graphical exploration and presentation scatterplots, barplots, boxplots, etc.
 Exporting tables for presentations in documents: cross tabulations regression, and other tables

Day 3—Statistical Analysis the R Way

Instructor: Dr. Paul Johnson, CRMDA Director and Professor, Political Science
 Regression & ANOVA
 Structural Equation Modeling
 Moderation and Mediation

Day 4—R Toolkit for Interacting with Data

Instructor: Dr. Paul Johnson, CRMDA Director and Professor, Political Science
 Matrix Algebra with R
 Iteration concepts in R: for, lapply
 Subsetting data, processing subsets and merging results
 Creating R functions to customize analysis

Day 5—Monte Carlo Programming and Power Analysis Instructors: Dr. Paul Johnson and Dr. Ben Kite

 Monte Carlo simulation in R
 Power analysis: definition and implications
 Using Monte Carlo simulation to estimate power

Day 6—An Introduction to Stata for Statistical Analysis

Instructor: Dr. Jacob Fowles, CRMDA and School of Public Affairs & Administration
 The Stata interface: point & click, menus, command line, and the do-file editor

  •  Finding help, including web sources
  •  Editing data within Stata, the project, data, and variable views
  •  Managing collaborative projects
  •  Reproducibility and documentation options within Stata

Day 7—Reliable and Reproducible Workflows Using Stata

Instructor: Dr. Jacob Fowles, CRMDA and School of Public Affairs & Administration
 Common syntax and structure for writing code

  •  Alternatives for editing do-files
  •  Workflow concepts, customized do-file template that facilitates project organization
  •  Importing, organizing, recoding, and labeling variables
  •  Generating descriptive plots and tables

Day 8—Automating Common Tasks in Stata

Instructor: Dr. Jacob Fowles, CRMDA and School of Public Affairs & Administration

  •  Installing and using prepared Stata packages
  •  Estimating quantitative models and capturing estimation output
  •  Stata macros
  •  Produce “pretty” output (summary statistics, regressions results, etc.), using the estout suite of command and putexcel commands

Day 9—Data Visualization in Stata

Instructor: Dr. Jacob Fowles, CRMDA and School of Public Affairs & Administration

  •  Programs for creating plots, charts, and graphs
  •  Customizing graphics using Stata’s suite of graphing commands
  •  Stata’s “margins” and “marginsplot” command for visualizing regression results

Day 10—Introduction to Structural Equation Modeling

Instructor: Dr. Edgar Merkle, University of Missouri, Department of Psychological Sciences

  •  Factor Analysis Overview
  •  Confirmatory Factor Analysis By Example
  •  Visualizing SEM with Path diagrams
  •  Diagnostics for Estimated Models

Day 11—Using Mplus and R Instructor: TBA

  •  Introducing the CRMDA SEM Code Repository
  •  Mplus
  •  Using R for basic SEM: the lavaan package

Day 12—Interaction Effects in SEM

Instructor: Dr. Holger Brandt, University of Kansas, Department of Psychology

  •  Lavaan overview and Product Indicators
  •  The R package nlsem
  •  Latent moderate structural equations (LMS)
  •  Graphical illustrations

Day 13—Extensions to semiparametric approaches

Instructor: Dr. Holger Brandt, University of Kansas, Department of Psychology

  •  Structural equation mixture modeling (SEMM)
  •  SEMM with the R packages nlsem and plotSEMM
  •  Robust alternatives for non-normality

Day 14—Bayesian alternatives and multilevel SEM

Instructors: Dr. Holger Brandt, University of Kansas, Department of Psychology

  •  (Short) introduction to Bayesian modeling
  •  Introduction to stan (a Bayesian analysis framework) and rstan (a R package for usage of stan)
  •  Interaction models with regression and the multilevel framework in stan
  •  Multilevel SEM with interaction effects in stan

For more information click "Further official information" below.


This opportunity has expired. It was originally published here:

http://crmda.ku.edu/statscamp



Similar Opportunities


Disciplines

Data Sciences

Statistics

Eligible Countries

International

Host Countries

United States