Examples
Examples
Try the starting example
- Baseline conversion fraction
- 0.1
- Users per equal-sized arm
- 10000
- Two-sided alpha
- 0.05
- Desired power
- 0.8
- Local normal MDE (absolute fraction)
- 0.011886
- MDE relative to baseline (%)
- 11.886119
Try another scenario
- Baseline conversion fraction
- 0.11000000000000001
- Users per equal-sized arm
- 10000
- Two-sided alpha
- 0.05
- Desired power
- 0.8
- Local normal MDE (absolute fraction)
- 0.012397
- MDE relative to baseline (%)
- 11.269833
Calculation method
Calculation method
Local normal MDE (absolute fraction)
Local null-variance approximation MDE=(z(1-alpha/2)+z(power))sqrt(2p(1-p)/n).
MDE relative to baseline (%)
Local null-variance approximation MDE=(z(1-alpha/2)+z(power))sqrt(2p(1-p)/n).
Symbols and units
- a
- Baseline conversion fraction
- b
- Users per equal-sized arm
- c
- Two-sided alpha
- d
- Desired power
Use the units shown. A list keeps the order in which you enter its values. The symbol ^ means “raised to the power of”.
Assumptions & limits
Assumptions & limits
- Planning approximation near the baseline, equal arms, no sequential peeking adjustment. Supported limits: All required fields must be finite values of the exact declared type and within displayed bounds. Lists contain1–1000 finite numbers. JSON tables are numeric rectangular rows, at most100 rows and30 columns; individual tools have stricter shape/domain constraints. Planning approximation near the baseline, equal arms, no sequential peeking adjustment.
Are these inputs private?
These calculations run in your browser. A shared link contains the values you choose to share. Only questions submitted to the separate AI finder are sent to OpenAI.
Further reading
Background reading; this does not imply endorsement or independent review of SumFizz.