Panel Local Indicators of Spatial Association (Panel LISA)
Panel Local Indicators of Spatial Association · Also known as: Panel LISA, spatiotemporal LISA, panel local spatial autocorrelation, LISA panel extension
Panel Local Indicators of Spatial Association extends Anselin's LISA statistics — most commonly Local Moran's I — to panel datasets, identifying spatial clusters and outliers at each location across multiple time periods. By applying local autocorrelation measures repeatedly over time, researchers can detect whether spatial concentration patterns emerge, persist, or dissolve, giving a richer spatiotemporal picture than a single cross-section allows.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
Use Panel LISA when you have georeferenced data measured repeatedly over time (e.g., annual regional statistics, repeated health surveys by district, property prices across quarters) and you want to identify where spatial clusters form, persist, or dissolve over time. It is well suited for exploratory spatial data analysis (ESDA) before fitting a spatial panel regression. Do not use it as a confirmatory inferential test on its own; it is primarily exploratory. It is not appropriate when the panel is very short (fewer than 3–4 periods), when spatial units are heterogeneous in size to a degree that distorts the weights matrix, or when the primary interest is global rather than local autocorrelation — in that case use Panel Moran's I instead.
Strengths & limitations
- Reveals spatially localised patterns that global statistics (e.g., global Moran's I) conceal, identifying which specific locations drive overall autocorrelation.
- Extends a well-validated single cross-section method to panel data, enabling the study of the temporal evolution of spatial clusters.
- Produces intuitive cluster maps (LISA maps) that communicate spatiotemporal patterns effectively to non-specialist audiences.
- Does not require a parametric distributional assumption; significance is assessed by permutation.
- Can serve as a diagnostic prior to spatial panel regression, guiding model specification and the choice of spatial lag versus error structure.
- Results are sensitive to the choice of spatial weights matrix; different neighbourhood definitions can produce markedly different cluster maps.
- Multiple testing is a serious concern when hundreds of locations are tested simultaneously; without correction, many false positives arise.
- The method is primarily exploratory and descriptive; it does not model causal relationships or control for confounders.
- Panel extension requires stable spatial units across time — mergers, splits, or boundary changes invalidate direct comparison.
- Permutation-based p-values are sensitive to the number of permutations and may differ slightly across software implementations.
Frequently asked
How does Panel LISA differ from running LISA separately for each year?
Running LISA separately for each year is exactly what Panel LISA does at the computation stage; the 'panel' extension lies in the comparative analysis across years — testing persistence, computing stability indices, or using the resulting local statistics as inputs to a panel regression. Using a consistent weights matrix and applying multiple-testing corrections across all periods together are best-practice panel-specific refinements.
Which weights matrix should I use?
Contiguity matrices (queen or rook) are common for polygonal areal units (e.g., administrative regions). Distance-band or k-nearest-neighbour matrices suit point or irregular data. The choice should be theoretically motivated by the diffusion or interaction mechanism you expect. Always test sensitivity to at least two alternative specifications and report differences.
How do I handle the multiple-testing problem?
Apply a false-discovery rate (FDR) correction (Benjamini-Hochberg) or a Bonferroni correction to the permutation p-values before mapping clusters. FDR is generally preferred for spatial data because it is less conservative. Some software packages (PySAL, spdep) offer built-in FDR corrections for LISA.
Can Panel LISA replace a spatial panel regression?
No. Panel LISA is exploratory — it identifies where spatial clusters exist over time but does not control for covariates or estimate structural parameters. Use it to motivate and specify a spatial panel model (SAR, SEM, SDM), not to replace one.
What if my panel is unbalanced (not all units observed in all periods)?
LISA can still be computed for each period independently using the units observed in that period, but direct comparison across periods becomes harder. If missingness is systematic rather than random, spatial clusters in underrepresented periods may reflect data gaps rather than true spatial patterns. Report the number of observations per period and check for spatial selectivity of missingness.
Sources
- Anselin, L. (1995). Local indicators of spatial association — LISA. Geographical Analysis, 27(2), 93–115. DOI: 10.1111/j.1538-4632.1995.tb00338.x ↗
- Millo, G., & Piras, G. (2012). splm: Spatial panel data models in R. Journal of Statistical Software, 47(1), 1–38. DOI: 10.18637/jss.v047.i01 ↗
How to cite this page
ScholarGate. (2026, June 3). Panel Local Indicators of Spatial Association. ScholarGate. https://scholargate.app/en/spatial-analysis/panel-local-indicators-of-spatial-association
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 Indicators of Spatial AssociationSpatial analysis↔ compare
- Local Moran's ISpatial analysis↔ compare
- Panel Spatial AutocorrelationSpatial analysis↔ compare
- Space-Time Local Indicators of Spatial AssociationSpatial analysis↔ compare