Hands-on training on Advanced Statistical Modelling using Stata and R

Description of the Course

This course aims to teach students to use advanced statistical modeling techniques and to interpret the results in the context of the research objectives. The training will focus on statistical models for Gaussian and non-Gaussian data in the context of health and related fields. These models include General Linear Models, Generalized Linear models (GLM), Linear Mixed Effect models (LMM), Generalized Linear Mixed Effect Models (GLMM), Multivariate Methods (PCA, FA, & SEM), semi parametric regression, nonlinear models, Survival data analysis methods (non-parametric methods, semi-parametric methods and parametric methods), Difference in Difference (DID), and Propensity Score matching (PSP). In addition, brief introduction will be given on Stata 16 for data management

• The trainees will be PhD students who are currently completed their data collection and ready for analysis
• It will be held at the University of Gondar

Content of the training

Chapter 1: Data Management Software

• General Introduction
• Introduction to STATA

Chapter 2: Introduction to Basic Statistical Models

• General Linear Models

• Generalized Linear Models

  • Multinomial Logistic regression
  • Poisson regression Model
  • Negative Binomial Regression

Chapter 3: Modeling Correlated Data

• Model for longitudinal Gaussian data

  • Linear Mixed Effect Model

• Models for correlated Non-Gaussian data (GLMM, GEE)

  • Generalized Estimating Equation
  • Generalized Linear Mixed Effect Model

• Missing Data and Management

  • Missing Mechanisms
  • Handling Missing Data

Chapter 4: Survival Data Analysis

• Nonparametric Methods
• Semiparametric Regression Model
• Parametric Models
• Frailty models

Chapter 5: Multivariate Methods

• Principal Components Analysis
• Factor Analysis
• Structural Equation Modeling

Chapter 6: Additional Models

• Difference-in-differences (DID)
• Propensity-Score Matching (PSM)
• Time-series Analysis

Chapter 7: Nonparametric methods

• Wilcoxon Rank-Sum Test
• Signed-rank test
• Mann Whitney U-test
• Kruskal-Wallis test

Tentative Schedule (December 22-29, 2022)

Topics	Date	Responsible
Chapter 1: Research Data Management	Day 1	Tadesse A, Malede M
- General Introduction
- Introduction to STATA
--------	-------	-----------------
Chapter 2:Basic statistical models	Day 2-3	Tadesse, A Malede M
- General Linear regression
- Generalized Linear Models
- Multinomial Logistic Regression
- Poisson Regression
--------	-------	-----------------
Chapter 3: Models for Correlated Data	Day 4-5	Tadesse A, Malede M
- Model for longitudinal Gaussian data
- Model for correlated Non-Gaussian data
- Generalized Estimating Equation (GEE)
- Generalized Linear Mixed Model (LMM)
- Missing mechanism
- Missing data handling
--------	-------	-----------------
Chapter 4: Survival Data Analysis	Day 6	Tadesse A, Malede M
- Introduction to survival analysis
- Nonparametric procedures
- Semiparametric modeling survival data
- Parametric survival models
- Computing risk models
- Frailty models
--------	-------	-----------------
Chapter 5: Multivariate Methods	Day 7	Tadesse A, Malede M
- Principal Components
- Factor Analysis
- Structural equation modelling
--------	-------	-----------------
Chapter 6: Other Models	Day 8	Tadesse A Malede M
-Difference-in-differences method (DID)
-Propensity-Score Matching (PSM)
-Time-series Analysis
--------	-------	-----------------
Chapter 7: Nonparametric methods	Day 8	Tadesse A, Malede M
--------	-------	-----------------

References

1. David Collett. Modeling survival data in medical research
2. David Collett. Binary data Modeling
3. Elisa T. Lee, John Wenyuwang. Statistical Methods for Survival Data Analysis
4. John Kloke: Joseph W. McKean: Nonparametric Statistical Methods Using R
5. Michael H. Kutner, Christopher J. Nachtsheim, John Neter and William Li: Applied Linear Statistical Models
6. Geert Molenberghs Geert Verbeke: Models for Discrete Longitudinal Data