# Floating population diversity as a leading indicator of business turnover imbalance in commercial districts: A spatial panel analysis in Seoul

## TL;DR

This study introduces a new way to predict commercial instability in Seoul by analyzing visitor diversity and business turnover patterns.

## Contribution

The study introduces a Commercial Instability Index and integrates floating population diversity with spatial analysis to predict commercial instability.

## Key findings

- Higher day-of-week diversity consistently reduces commercial instability.
- Age diversity partially mitigates instability in commercial districts.
- Commercial instability shows spatial dependence, spreading from one district to its neighbors.

## Abstract

This study develops a quantitative indicator for the early diagnosis of commercial instability in business openings and closures and proposes a new analytical framework for assessing the stability of commercial districts. To address the limitations of prior research that relied on static measures such as sales or store counts, two complementary approaches were introduced. First, we propose the Commercial Instability Index (CII), a direction-agnostic metric of turnover instability computed from the standardized relative deviation between openings and closures, where larger values indicate greater instability. Second, entropy-based floating population diversity indicators were applied to capture the distribution of visitors by age, time of day, and day of week, as well as their temporal changes. These indicators were tested on quarterly panel data from 1,650 commercial districts in Seoul between the first quarter of 2019 and the fourth quarter of 2024 using panel regression and spatial panel regression models, specifically the spatial autoregressive (SAR) and spatial error model (SEM). The results showed that higher day-of-week diversity consistently reduced the CII, whereas a greater balance in age diversity provided partial mitigation effects. Moreover, the CII revealed significant spatial dependence, indicating that instability in one district could spread to its neighbors. By focusing on the magnitude of instability rather than its direction, and by integrating floating population diversity with spatial dependence, this study advances beyond static approaches. These findings expand the theoretical scope of commercial district research and offer a practical basis for early warning systems and area-based management strategies, thereby contributing to the development of urban policies for resilience and stability of commercial districts.