Function recommendSampleSize

recommendSampleSize(
    opts: {
        confidenceLevel?: number;
        desiredMarginOfError?: number;
        minDetectableEffect?: number;
        pilotSD?: number;
        power?: number;
    },
): SampleSizeRecommendation
Recommend sample size for a desired margin of error on a single score.

Uses the formula: n = (z / E)^2 * s^2 where z is the critical value, E is margin of error, s is estimated SD.

If no pilot SD is provided, uses 0.25 (conservative for [0,1] bounded scores).
Parameters
- opts: {
      confidenceLevel?: number;
      desiredMarginOfError?: number;
      minDetectableEffect?: number;
      pilotSD?: number;
      power?: number;
  }
  - OptionalconfidenceLevel?: number
  - OptionaldesiredMarginOfError?: number
  - OptionalminDetectableEffect?: number
    
    For comparison designs: minimum detectable effect size (Cohen's d)
  - OptionalpilotSD?: number
  - Optionalpower?: number
    
    For comparison designs: desired power (default 0.80)
Returns SampleSizeRecommendation
- Defined in src/score/statisticalAnalysis.ts:339