statistical-analysis

Compute descriptive statistics and detect outliers in numeric datasets.

1|Updated Feb 19, 2026
One-click install
npx skills add https://github.com/cyyeh/duckdb-data-agent --skill statistical-analysis-cyyeh
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: statistical-analysis
Source: https://github.com/cyyeh/duckdb-data-agent/tree/main/plugins/data/skills/statistical-analysis
Command: npx skills add https://github.com/cyyeh/duckdb-data-agent --skill statistical-analysis-cyyeh

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Provides clear, practical guidance to summarize numeric data, identify trends and anomalies, and evaluate whether observed differences are statistically meaningful so non-expert analysts can make defensible decisions from datasets.

Core Features & Use Cases

  • Descriptive statistics: recommended measures of center and spread (mean, median, IQR, standard deviation) and relevant percentiles for business reporting.
  • Trend analysis and simple forecasting: moving averages, period-over-period comparisons, seasonality checks, and naïve/seasonal forecasts for short-term planning.
  • Outlier and anomaly detection: z-score, IQR, and percentile methods with investigation and handling guidance rather than blind removal.
  • Hypothesis testing guidance: when to apply t-tests, chi-squared tests, ANOVA, nonparametric alternatives, plus advice on effect sizes, confidence intervals, and sample size considerations.
  • Use cases: A/B test evaluation, monthly retention trend summaries, fraud/outlier investigation in transaction data, and executive-ready summaries that balance statistical and practical significance.

Quick Start

Use the statistical-analysis skill to compute descriptive summaries, detect outliers, and run an appropriate hypothesis test on the uploaded dataset to support a business decision.

Frequently Asked Questions about statistical-analysis

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I detect outliers in a CSV dataset using z-score or IQR methods?

Outlier detection in a CSV dataset applies z-score and IQR methods to numeric columns to flag anomalies, providing investigation guidance rather than blind removal. You supply numeric columns and optional grouping variables to receive outlier flags.

What statistical test should I use for A/B test evaluation?

A/B test evaluation uses hypothesis testing guidance to determine when to apply t-tests, chi-squared tests, or ANOVA based on your data structure. Results include p-values, effect sizes, and confidence intervals to balance statistical and practical significance.

How do I compute descriptive statistics for business reporting from database tables?

Descriptive statistics for database-extracted tables compute measures of center and spread including mean, median, IQR, and standard deviation with relevant percentiles. You provide numeric columns and optional grouping variables to generate distributional summaries.

Can I perform time series trend analysis and forecasting on Parquet files?

Time series trend analysis on Parquet files computes moving averages, period-over-period comparisons, and seasonality checks with naïve or seasonal forecasts. This supports short-term planning by applying trend smoothing to your numeric datasets.

When should I use nonparametric alternatives instead of standard hypothesis tests?

Nonparametric alternatives apply when your data violates normality assumptions required by standard t-tests or ANOVA. The skill advises on appropriate test selection, effect sizes, confidence intervals, and sample size considerations for defensible decisions.

Does this statistical analysis approach work with grouped numeric data?

Statistical analysis works with grouped numeric data by accepting optional grouping variables alongside numeric columns. This enables segmented descriptive statistics, outlier detection, and hypothesis testing across distinct categories in your dataset.