[ecdf-basic] Basic ECDF Plot (Empirical Cumulative Distribution)

[MarkusNeusinger]

Description

An ECDF (Empirical Cumulative Distribution Function) plot shows the proportion of observations less than or equal to each value. It displays a step function that increases by 1/n at each data point, providing a non-parametric estimate of the cumulative distribution.

Key Features

• Step function showing cumulative probability
• No binning or smoothing required (unlike histograms)
• Y-axis ranges from 0 to 1 (proportion of data)
• Direct reading of percentiles and quantiles

Applications

• Comparing distributions between groups
• Identifying distribution characteristics (median, quartiles)
• Statistical analysis and hypothesis testing
• Visualizing sample distribution without parametric assumptions

Data Requirements

• values (numeric) - Continuous variable
• Size: Any sample size (works well from 10 to 10000+ observations)

Visual Elements

• Step function (stairs) increasing at each observation
• Y-axis from 0 to 1 representing cumulative proportion
• Optional: confidence bands
• Optional: multiple ECDF lines for group comparison

[claude]

✅ Spec ID: ecdf-basic

Valid plot type. ECDF is a distinct statistical visualization for cumulative distributions. No duplicates exist - this complements histogram-basic and density-basic as another distribution visualization approach.

Add label approved to begin generation.

[github-actions]

✅ Specification Ready: ecdf-basic

PR: #993

---

specification.md

ecdf-basic: Basic ECDF Plot

Description

An ECDF (Empirical Cumulative Distribution Function) plot displays a step function that shows the proportion of observations less than or equal to each value. Unlike histograms, ECDF plots require no binning or smoothing, providing a non-parametric estimate of the cumulative distribution. The y-axis ranges from 0 to 1, allowing direct reading of percentiles and quantiles from the visualization.

Applications

• Comparing distributions between groups to identify differences in spread, location, or shape
• Identifying distribution characteristics such as median, quartiles, and percentiles at a glance
• Statistical analysis and hypothesis testing where the full distribution shape matters
• Visualizing sample distributions without making parametric assumptions about the underlying data

Data

• values (numeric) - Continuous variable to visualize
• Size: 50-500 observations (works well from 10 to 10000+ but 50-500 is ideal for visualization)
• Example: Random samples from a normal distribution

Notes

• The step function should increase by 1/n at each data point
• Y-axis must range from 0 to 1 representing cumulative proportion
• Consider using a distinct line style for clarity
• Grid lines help with reading specific percentile values

---

Next: Add approved label to merge this specification to main.

---

🤖 spec-create workflow