This page demonstrates kernel density estimation for visualizing continuous distributions.
Smooth density curves: Non-parametric estimation of probability density
Multiple overlapping densities: Compare distributions across groups with transparency
Interactive filters: Slider controls for categorical and numeric ranges
Faceting: Split visualizations by one or two categorical variables
Use sliders and multi-select controls to filter and explore the data!
Simple Kernel Density
Basic kernel density estimation for a normal distribution
auto(0 = auto, max ≈ 7.9)
Data: df1
Treatment Comparison - Overlapping Densities
Compare distributions with overlapping density curves. Transparency allows visibility of all groups.
auto(0 = auto, max ≈ 5.9)
Data: df2
Score Distribution by Department with Filters
Use age range slider and region multi-select to filter data dynamically and see how distributions change
Filters
18.0 to 80.0
Hold Ctrl/Cmd to select multiple values
auto(0 = auto, max ≈ 11.4)
Data: df3
Faceted Kernel Density - Single Variable
Faceting by phase shows separate density plots for each phase, with categories overlaid within each facet
Faceting
auto(0 = auto, max ≈ 7.9)
Data: df4
Faceted Grid - Two Variables
Two-way faceting creates a grid showing all combinations of treatment and timepoint
Faceting
auto(0 = auto, max ≈ 9.2)
Data: df5
Bimodal Distribution Detection
Kernel density estimation excels at revealing complex distribution shapes like bimodality
auto(0 = auto, max ≈ 13.0)
Data: df6
Custom Bandwidth Setting
Bandwidth controls smoothness: smaller values show more detail, larger values create smoother curves
2.0(0 = auto, max ≈ 3.9)
Data: df7
Multi-Variable KDE with Dropdowns and Bandwidth Control
Select different variables and grouping columns using the dropdowns. Adjust bandwidth to control smoothness. Set bandwidth to 0 for automatic calculation.
Plot Attributes
auto(0 = auto, max ≈ 39.2)
Data: df8
Comprehensive Example - All Features Combined
This example demonstrates all KernelDensity features together: multiple value columns, grouping by color, filtering with sliders, and faceting. Use the controls to explore different combinations and see how distributions vary across departments, experience levels, and project types.
Filters
20.0 to 65.0
Hold Ctrl/Cmd to select multiple values
Plot Attributes
Faceting
auto(0 = auto, max ≈ 13.1)
Data: df9
Key Features Summary
Smooth density estimation: Non-parametric visualization of continuous distributions
Overlapping groups: Compare multiple distributions with transparency for visibility
Interactive filtering: Range sliders for numeric columns, multi-select for categorical
Flexible faceting: Split plots by one or two categorical variables
Bandwidth control: Interactive input to adjust smoothness (0 for automatic using Silverman's rule)
Variable selection: Dropdowns to switch between multiple value and group columns
Bimodality detection: Reveals complex distribution shapes that histograms might miss
Tip: Kernel density plots are particularly useful for comparing continuous distributions and detecting complex patterns!