Analysis

Latest Post

5 May, 2021

Causal inference: Inverse probability treatment weighting

The tutorial is based on R and StatsNotebook, a graphical interface for R. In multiwave longitudinal study, the exposure is often time-varying. A time varying confounder is a time varying variable that is affected by previous exposures, and also affect future exposure and outcome, thus confounding the relationship between the exposure and the outcome. Time varying confounding is common in longitudinal research. When evaluating the relationship between a time-varying exposure and outcome, standard methods such as regression and generalised linear models produce biased estimates in the presence of time-varying confounders. Read more

All Posts

  • Causal inference: Inverse probability treatment weighting - 5 May, 2021
  • Age-period-cohort analysis - 29 Mar, 2021
  • Robust regression - 8 Dec, 2020
  • IPTW with missing data - 5 Dec, 2020
  • Linear mixed model (Multilevel model) - 20 Oct, 2020
  • Residual plots and assumption checking - 16 Oct, 2020
  • Multiple Imputation - 23 Sep, 2020
  • Causal Mediation Analysis - 23 Sep, 2020
  • Moderation (Interaction) analysis using linear regression - 22 Sep, 2020
  • Descriptive statistics - 20 Sep, 2020
  • Linear regression with missing data - 18 Sep, 2020
  • Linear Regression - 18 Sep, 2020