CLI: Sensitivity¶

causal-toolkit sensitivity --data DATA.csv --treatment T --outcome Y [OPTIONS]

Options¶

Option	Short	Description
`--data`	`-d`	Input CSV
`--treatment`	`-t`	Treatment column
`--outcome`	`-y`	Outcome column
`--confounders`	`-c`	Confounder columns
`--method`	`-m`	`rosenbaum`, `cinelli_hazlett`, `evalue`, `all` (default: `all`)
`--estimate`	`-e`	Point estimate (optional, runs estimation if omitted)
`--se`		Standard error (optional)
`--out`	`-o`	Output directory (default: `./sensitivity`)

Examples¶

# Full suite (runs estimation first if --estimate/--se not provided)
causal-toolkit sensitivity -d data.csv -t treatment -y outcome -c age income education

# Just E-value with known estimate
causal-toolkit sensitivity -d data.csv -t treatment -y outcome \
    --estimate 2.5 --se 0.45 -m evalue

Output¶

Rosenbaum bounds: Γ=1.84, conclusion_reversed=True
Cinelli-Hazlett: RV=0.12, R²_yz=0.15, R²_zd=0.08
E-value: 2.45 (CI: 1.67)
  Benchmark 'income': R²_yz=0.08, R²_zd=0.05
  Benchmark 'education': R²_yz=0.03, R²_zd=0.02
[green]Sensitivity results saved to sensitivity/sensitivity.json[/green]