Factor Analysis for Land Value Index in Urban Areas Using Agent Analysis Indicator

Ota, Akira and Kobayashi, Junya and Kaneda, Toshiyuki (2022) Factor Analysis for Land Value Index in Urban Areas Using Agent Analysis Indicator. Mobility, Knowledge and Innovation Hubs in Urban and Regional Development. Proceedings of REAL CORP 2022, 27th International Conference on Urban Development, Regional Planning and Information Society. pp. 365-373. ISSN 2521-3938

Text (Factor Analysis for Land Value Index in Urban Areas Using Agent Analysis Indicator) CORP2022_60.pdf - Published Version Download (699kB)

Abstract

The factors affecting a land value index such as land assessments are important for the development and growth of urban areas. Ota and Kaneda (2018) conducted a comparative analysis of a land value index in the central Nagoya area of Japan and reported that the factor structure could be explained by three factors: distance from the nearest station as an accessibility factor; the concentration of neighborhood commercial and business uses as a facility volume factor; and the integration value of the entire area as an indicator of the street network centrality of the vis graph analysis of space syntax theory, or “VGA”, as a space configuration factor. In an urban area, a busy street’s land value index is considered to be higher. The integration value of the VGA indicator, which represents the street network centrality as a space configuration, has been used as a busy street factor. However, high street network centrality is not always needed for a busy street. Therefore, it is a possible that simulating actual pedestrians from the space configuration is a stronger factor for a busy street than a high street network centrality. Simulating actual pedestrians from the space configuration can be conducted using agent analysis, or “AA.” In this paper, we examine a multiple regression model for the factors and a land value index of the Kanayama area of Nagoya City using a VGA indicator and then replacing the VGA indicator with the AA indicator as a new factor. By comparing the two models, we explore the potential for using the AA indicator as a land value index factor. In conclusion, the global integration value of the VGA indicator was selected as a factor for a busy street with a multiple correlation coefficient of 0.750, a coefficient of determination of 0.562, and an Akaike information criterion (AIC) of 352.093 with a standard partial regression coefficient of 0.362 in the conventional factor structure. On the other hand, when the number of AA footprints (station occurrences) of the AA indicator was selected as a factor for a busy street, it had a multiple correlation coefficient of 0.830, a coefficient of determination of 0.689, and an AIC of 294.477 with a standard partial regression coefficient of 0.618 in the new factor structure. Thus, we discovered that replacing the VGA indicator with the AA indicator could significantly improve the land value factor structure model.

Item Type:	Article
Uncontrolled Keywords:	space syntax, space configuration, land value, visibility graph analysis, agent analysis
Subjects:	G Geography. Anthropology. Recreation > GB Physical geography H Social Sciences > HA Statistics H Social Sciences > HD Industries. Land use. Labor
Depositing User:	The CORP Team
Date Deposited:	10 Nov 2022 17:53
Last Modified:	18 Dec 2022 14:11
URI:	http://repository.corp.at/id/eprint/900

Actions (login required)

View Item