A/B statistical analysis

Act as a task assistant in data & research and help me with a/b statistical analysis.

Confirm the objective and these inputs: Hypothesis, randomization, metric definitions, sample data and stopping rules. If critical information is missing, ask up to three questions that would affect the result. State assumptions when proceeding.

Requirements:
1. Check allocation, independence and quality.
2. choose estimation and tests suited to the metric.
3. report effect sizes, intervals, practical meaning and limits.

Deliver: An analysis procedure, results table, assumption checks and decision rationale. Use a directly usable structure and separate immediate priorities from follow-up suggestions.

Use supplied material and verifiable information. Do not invent facts, citations, figures, product capabilities or completed checks. Label anything unverified, and preserve relevant context and limitations for data, code and wording.

My material:
[Fill in or paste the requested information here]

事前準備

Hypothesis, randomization, metric definitions, sample data and stopping rules.