Executive summary: A cost effectiveness analysis using interval estimates for inputs can greatly overestimate uncertainty. Modeling inputs with probability distributions instead leads to more accurate uncertainty estimates and enables additional types of inference.
Key points:
Propagating interval estimates through a model compounds uncertainty and produces very wide bounds on outputs. Using probability distributions for inputs gives more accurate uncertainty estimates.
The tool Distributr allows exploring different distributions to match uncertain estimates and generates samples for spreadsheet calculations.
A probabilistic analysis of GiveWell’s StrongMinds evaluation shows much less uncertainty than an interval-based analysis.
Probabilistic modeling enables comparing interventions via distributions and visualizations not possible with intervals.
Incorporating uncertainty into cost effectiveness analyses can be done without coding by combining Distributr with spreadsheets.
Switching from intervals to probability distributions increases validity of inferences and opportunities for nuanced comparisons.
This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, andcontact us if you have feedback.
Executive summary: A cost effectiveness analysis using interval estimates for inputs can greatly overestimate uncertainty. Modeling inputs with probability distributions instead leads to more accurate uncertainty estimates and enables additional types of inference.
Key points:
Propagating interval estimates through a model compounds uncertainty and produces very wide bounds on outputs. Using probability distributions for inputs gives more accurate uncertainty estimates.
The tool Distributr allows exploring different distributions to match uncertain estimates and generates samples for spreadsheet calculations.
A probabilistic analysis of GiveWell’s StrongMinds evaluation shows much less uncertainty than an interval-based analysis.
Probabilistic modeling enables comparing interventions via distributions and visualizations not possible with intervals.
Incorporating uncertainty into cost effectiveness analyses can be done without coding by combining Distributr with spreadsheets.
Switching from intervals to probability distributions increases validity of inferences and opportunities for nuanced comparisons.
This comment was auto-generated by the EA Forum Team. Feel free to point out issues with this summary by replying to the comment, and contact us if you have feedback.