Skip to contentScholarGate
LibraryBookshelfDeskReview StudioAssistant
Sign in
On this page
IntuitionHow it worksWhen to use itStrengths & limitationsCommon pitfallsApplicationsFrequently asked🔒 Read the full methodSourcesRelated methods
Cite this pageSpotted an issue on this page? Report or suggest a fix →
Home›Spatial analysis›Panel Kernel Density Estimation
Regression modelGIS / spatial

Panel Kernel Density Estimation

Also known as: Panel KDE, longitudinal kernel density estimation, repeated-measures KDE, panel nonparametric density estimation

Panel Kernel Density Estimation (Panel KDE) extends the standard kernel density estimator to panel (longitudinal) data, estimating smooth density surfaces for spatial or attribute variables observed across multiple units and time periods. It reveals how the distribution of a phenomenon shifts, concentrates, or disperses over time and across groups, making it a natural tool for tracking spatial patterns in repeated-measures or panel datasets.

ScholarGate
  1. Regression model
  2. v1
  3. 2 Sources
  4. PUBLISHED
Cite this page →
Tools & resources
Download slides
Learn & explore

Read the full method

Members only

Sign in with a free account to read this section.

Sign in

Method map

The neighbourhood of related methods — select a node to explore.

Panel Kernel Density Estimation
Local Kernel Density Est…Panel Hot Spot AnalysisPanel Spatial Autocorrel…Panel Spatial RegressionSpace-Time Kernel Densit…

When to use it

Use Panel KDE when you have a spatial or continuous variable measured repeatedly across the same units or locations and you want to track how the probability density (concentration of activity) evolves over time or across groups — for example crime counts in city districts over years, air-pollutant concentrations at monitoring stations across seasons, or house-price distributions across urban sub-markets over decades. Panel KDE is ideal when no parametric distributional assumption is warranted and the goal is exploratory or visual. Do not use it when you need causal inference or regression coefficients; when the panel is very short (T < 3) the temporal comparison is trivial; or when observations per period are too few (n < 30 per slice) to support a reliable density estimate.

Strengths & limitations

Strengths
  • Fully nonparametric: no distributional assumption is imposed on the data within each panel slice.
  • Produces an interpretable, continuous density surface that can be overlaid on maps or compared visually across periods.
  • Sensitive to clustering, multi-modality, and shifting hotspots that parametric summaries (means, variances) would miss.
  • Flexible pooling strategies allow borrowing strength across sparse panels while retaining period-specific estimates.
  • Widely implemented in spatial statistics software (R: spatstat, ks; Python: scipy, KDEpy; ArcGIS Kernel Density tool).
Limitations
  • Bandwidth choice strongly influences the result; there is no universally optimal selector for panel data, and naive use of Silverman's rule can over-smooth multimodal distributions.
  • Inference (formal testing of distributional change across periods) requires bootstrap or permutation procedures that add computational cost.
  • Edge effects at the boundary of the study area bias density estimates unless boundary correction is applied.
  • Does not produce regression coefficients or causal estimates; it is descriptive, not explanatory.
  • Computationally intensive for very large spatial panels (millions of observations per period).

Frequently asked

How is Panel KDE different from standard KDE?

Standard KDE produces a single density surface from one sample. Panel KDE applies KDE repeatedly to data organised by panel index (time period, group, or both), producing a sequence or set of surfaces that can be compared. The core estimator is the same; the added complexity lies in bandwidth harmonisation and temporal comparison.

How do I choose the bandwidth for panel data?

A common approach is to select a single bandwidth via cross-validation on the pooled dataset and apply it to all panels so the surfaces are comparable. Alternatively, adaptive bandwidths that adjust to local density can be applied per panel. Avoid using the default rule-of-thumb if your distributions are multimodal or if panel sizes differ substantially.

Can I formally test whether the density has changed between periods?

Yes. Bootstrap-based tests of the integrated squared difference between two density surfaces (f_t and f_{t-1}) provide a formal significance test. Permutation approaches that shuffle period labels also work. Software packages such as R's ks and spatstat provide these routines.

What if some panels have very few observations?

Sparse panels produce unreliable density estimates even with optimal bandwidths. Consider pooling adjacent periods, using an adaptive bandwidth that widens in low-density regions, or applying a mixed-effects density smoother that borrows strength from other panels.

Is Panel KDE the same as space-time KDE?

They overlap but differ in emphasis. Space-time KDE simultaneously smooths over both geographic space and time using a single joint kernel, treating the time dimension as a continuous axis. Panel KDE treats time as discrete panel index and produces separate spatial density surfaces per period, which are then compared. Panel KDE is more natural when periods correspond to discrete policy regimes, years, or survey waves.

Sources

  1. Parzen, E. (1962). On estimation of a probability density function and mode. Annals of Mathematical Statistics, 33(3), 1065-1076. DOI: 10.1214/aoms/1177704472 ↗
  2. Silverman, B. W. (1986). Density Estimation for Statistics and Data Analysis. Chapman and Hall, London. ISBN: 978-0412246203

How to cite this page

ScholarGate. (2026, June 3). Panel Kernel Density Estimation. ScholarGate. https://scholargate.app/en/spatial-analysis/panel-kernel-density-estimation

Related methods

Local Kernel Density EstimationPanel Hot Spot AnalysisPanel Spatial AutocorrelationPanel Spatial RegressionSpace-Time Kernel Density Estimation

Which method?

Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.

  • Local Kernel Density EstimationSpatial analysis↔ compare
  • Panel Hot Spot AnalysisSpatial analysis↔ compare
  • Panel Spatial AutocorrelationSpatial analysis↔ compare
  • Panel Spatial RegressionSpatial analysis↔ compare
  • Space-Time Kernel Density EstimationSpatial analysis↔ compare
Compare side by side →

Similar methods

Space-Time Kernel Density EstimationPanel KrigingLocal Kernel Density EstimationBayesian Kernel Density EstimationPanel Geographically Weighted RegressionPanel Ordinary KrigingPanel Universal KrigingPanel Local Indicators of Spatial Association

Related reference concepts

Density EstimationSpatial Point ProcessesNonparametric StatisticsPanel Data Models • Spatio-temporal ModelsPanel Data Models • Spatio-temporal ModelsEconometrics

Spotted an issue on this page? Report or suggest a fix →

ScholarGate — Panel Kernel Density Estimation (Panel Kernel Density Estimation). Retrieved 2026-07-20 from https://scholargate.app/en/spatial-analysis/panel-kernel-density-estimation · Dataset: https://doi.org/10.5281/zenodo.20539026
Quick facts
Originator
Parzen (1962); Silverman (1986); extended to panel contexts in spatial econometrics literature
Year
1962 (KDE); panel extension: 1990s–2000s
Type
Nonparametric density estimation
DataType
Continuous or count spatial data observed across multiple units and time periods (panel structure)
Subfamily
GIS / spatial
Related methods
Local Kernel Density EstimationPanel Hot Spot AnalysisPanel Spatial AutocorrelationPanel Spatial RegressionSpace-Time Kernel Density Estimation
ScholarGate

A content-first reference library for research methods — what each one is, how it works, and where it comes from.

Open data (CC-BY)

Explore

  • Library
  • Search the library…
  • Browse by field
  • Fields
  • Journey
  • Compare
  • Which method?

Reference

  • Subjects
  • Atlas
  • Glossary
  • Methodology
  • Philosophy

Your tools

  • Bookshelf
  • Desk
  • Chat

Company

  • About
  • Pricing
  • Contact
  • Suggest a method

Entries are compiled from published sources for reference. Verifying the accuracy and suitability of any information for your own use remains your responsibility.

© 2026 ScholarGate · A research-method reference library
  • Privacy
  • Cookies
  • Terms
  • Delete account