Search

Opening a new train station is considered a way to generate amenity in a neighborhood. However, as train lines extend from a central city to suburbs and remote places, train stations may generate disamenity depending on the local context. This study examines the externality of stations from a house valuation perspective. A mixed-effects model is employed to capture the varying relationship between house prices and distance to a nearby station. The results show that this relationship significantly varies by county, which leads to a house price premium in some counties and price discount in others. This study attributes the price discount (disamenity) to low ridership, seasonality of ridership, passenger traits, and long distance from a central city. The study results are expected to provide policymakers with balanced insights on establishing a new station so that train stations can serve as a local amenity, not a harmful facility in the neighborhood. [ABSTRACT FROM AUTHOR]

Using point-of-sales (POS) data, the sales trends of 48 member stores of a Korean restaurant franchise during the COVID-19 pandemic were analyzed. As daily sales are nested in each member store of a franchise, the hierarchical structure of POS data was fully and effectively utilized by employing a mixed-effects model. The results showed that although sales volumes in all member stores were negatively affected by the pandemic, the level of impact varied according to store location: sales at some stores were drastically reduced, while a few others even achieved a slight increase in sales during the pandemic. These findings suggest that the government support policy for small business owners should be designed in a locally optimized way, to take account of neighborhood characteristics and the degree of sales loss for individual business owners. [ABSTRACT FROM AUTHOR]

Abandoned houses have become a common feature of the local landscapes: the rising number of abandoned houses is a major challenge facing many counties in South Korea. Their presence negatively influences the neighborhood by undermining its aesthetic quality, depreciating the perception of safety in the neighborhood properties, and deepening the fiscal deficit of local financing. The detection of abandoned houses is the first step toward adequate housing management by local governments. This study aims to provide a cost-effective and prompt approach to identifying abandoned houses in rural areas. Multi-source data, that is, images and building registry data are utilized and a multi-input neural network is designed to adopt these heterogeneous datasets. Trained by the two source datasets, the proposed network achieves 86.2% accuracy in classifying abandoned houses, which is an acceptable performance level in administrative practice. The database of abandoned houses identified in this manner is expected to promote effective housing management by governments and ultimately contribute to mitigating vacancies in rural areas. [ABSTRACT FROM AUTHOR]

In contrast to the outstanding performance of the machine learning approach, its adoption in industry appears to be relatively slow compared to the speed of its proliferation in a variety of business sectors. The low interpretability of a black-box-type model, such as a machine learning-based valuation model, is one reason for this. In this study, house prices in Seoul and Jeollanam Province, South Korea, were estimated using a neural network, a representative model to implement machine learning, and we attempted to interpret the resultant price estimations using an interpretability tool called a partial dependence plot. Partial dependence analysis indicated that locally optimized valuation models should be designed to enhance valuation accuracy: a land-oriented model for Seoul and a building-focused model for the Jeollanam Province. The interpretable machine learning approach is expected to catalyze the adoption of machine learning in the industry, including property valuation. [ABSTRACT FROM AUTHOR]

The setting of farmland prices in the market not only reflects existing agricultural activities but also expected potential for development. This study decomposes farmland prices into values representative of current agricultural production and the prospective development potential at the county level in South Korea. The income value of farmland is derived by analysing agricultural revenue and production cost, and the sale value of farmland is estimated by reviewing transaction prices filed with the administrative authority. The difference between income value and sale value is adopted as the development value in this study. The results of the estimation show that the proportion of development value in the price of farmland is remarkably high, with a median proportion of 0.78, indicating that the threat of converting land to non-agricultural use is non-trivial because it remains a financially attractive alternative. In addition, the magnitude of the portion of the development value in the price of farmland varies considerably across counties depending on the distance to nearby metropolitan cities. This implies that agricultural policy should be designed in a locally optimised manner to effectively restrain the conversion of farmland for urban use. [ABSTRACT FROM AUTHOR]

Although deep learning-based valuation models are spreading throughout the real estate industry following the artificial intelligence boom, property owners and investors continue to doubt the accuracy of the results. In this study, we specify a neural network for predicting house prices. We suggest a standard feed-forward network with two hidden layers, and show that it is sufficiently reasonable to apply its prediction to real-world projects such as property valuation. In addition, we propose a Bayesian neural network for describing uncertainty in house price predictions while providing a means to quantify uncertainty for each prediction. We choose Gangnam-gu, Seoul for the analysis, and predict house prices in the area using both networks. Although the Bayesian neural network did not perform better than the conventional network, it could provide a tool to measure the uncertainty inherent in predicted prices. The findings of this study show that a Bayesian approach can model uncertainty in property valuation, thereby promoting the adoption of deep learning tools in the real estate industry. [ABSTRACT FROM AUTHOR]