statistical-analysis

To calculate effect sizes and confidence intervals for experimental data, this skill computes Cohen's d and applies bootstrap confidence intervals. It uses standard scientific tooling like numpy and scipy to quantify uncertainty and provide interpretable metrics for your results.

To run statistical hypothesis tests for method comparisons, this skill applies paired t-tests and non-parametric tests. It determines statistical significance and applies multiple comparisons corrections to ensure rigorous analysis across experimental ablations.

Yes, you can perform a power analysis for sample size planning in Python. This skill handles study design considerations to calculate required sample sizes for future experiments, relying on standard scientific computing libraries like scipy to execute the analysis.

The best way to generate publication-ready statistical summaries from experiment results is to synthesize p-values, confidence intervals, and effect sizes together. This skill processes experimental outputs to produce rigorous reports suitable for publication.

You need standard scientific Python tooling such as numpy and scipy to compute bootstrap confidence intervals and t-tests. These libraries provide the computational foundation required to execute the statistical analyses and power calculations.

You should use non-parametric tests instead of paired t-tests for ablation analysis when your experimental data does not meet normality assumptions. This skill supports both testing approaches, allowing you to choose the appropriate method to determine significance.