Long-term causal effects of recommendation exposure

Causal Inference
Longitudinal Data
Marginal Structural Models
G-Computation
A sequential causal project estimating whether short-term recommendation exposure affects longer-term engagement and retention-style outcomes.

Decision Question

How do sequential recommendation exposures affect future engagement or retention?

Causal Setup

  • Treatment is short-term recommendation exposure across time.
  • Outcome is future engagement and retention-style behavior.
  • The main design challenge is time-varying confounding affected by prior exposure.

Methods

  • Long-term outcome definition
  • Time-dependent propensity weights
  • Marginal structural models
  • G-computation
  • Doubly robust and heterogeneous effect analysis

Decision Takeaway

The project shows that durable value often requires sequential estimands, time-varying diagnostics, and longer-horizon outcome definitions.

Selected Figures

01 Estimator Comparison

02 Weight Balance

04 Secondary Outcomes

Notebook Sequence

  1. Notebook 01: KuaiRec Sequence EDA for Long-Term Causal Effects
  2. Notebook 02: Defining the Long-Term Causal Estimand
  3. Notebook 03: Time-Varying Confounding and Propensity Weights
  4. Notebook 04: Marginal Structural Model for Long-Term Effects
  5. Notebook 05: G-Computation for Long-Term Effects
  6. Notebook 06: Doubly Robust Heterogeneous Effects

Generated Artifacts

Limitations

These notebook-driven causal analyses should be read with their identification assumptions, support diagnostics, measurement choices, and sensitivity checks in view.