IMPLEMENTATION OF SPATIAL PLANNING ZONE AS A DETERMINING VARIABLE OF LAND PRICE ESTIMATION BASED ON GEOGRAPHICALLY WEIGHTED REGRESSION (GWR) (CASE STUDY: EASTERN BANDUNG CITY)

• Main Author: Nashrul Jabbaaar, Dzikri
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/66839
• Institution: Institut Teknologi Bandung

The variability of land prices is an implication of the existence of basic human needs, especially in the economic aspect. An understanding of land prices is needed to answer this problem especially related to land price estimation. In this study, the method used to estimate land prices is Geographically Weighted Regression (GWR). This method can find the relationship between the impact of driving factors and variations in spatial relationships. Generally, the GWR method is built based on the dependent variable (land prices) and independent (the spatial proximity between the land object and public facilities). However, in this study the independent variable will be developed by adding a spatial planning zone to provide the complexity of determining land price estimates based on spatial planning regulations, namely Bandung City Regional Regulation No. 18 of 2011. This regulation will be used as the basis for determining spatial planning zone variables for modeling land price estimates in the Eastern Bandung City. The implementation of spatial planning zones as a determinant of land price estimation is expected to form a better GWR model compared to models that do not involve spatial planning zone variables. This study uses 15 variables with 10 physical object variables in the form of public facilities and 5 variables in the form of spatial planning zones which have a significant influence on the pattern and distribution of land prices in the Eastern Bandung City. The fifteen variables are worship, industry, government offices, health, sports/recreation, education, prisons (correctional institutions), defense offices, terminals, trade and service zones, industrial zones, low settlement zones, medium settlement zones, and high settlement zones. To form the GWR model for land price estimation, 4 combinations of sample points and test points were used in two types of models (with spatial planning zones or not involving these variables). In this combination there is a terminology for determining outliers at both the sample and test points. This is done as a form of identification of anomalies from the data used in the hope that it can make different variations in each combination formed. This will also relate to the determination of bandwidth to determine the use of data points to solve the GWR equation. The results of this study indicate that the implementation of the spatial planning zone variable to form a model of land price estimation based on GWR in addition to the object variable of public facilities can provide a better level of accuracy than the GWR model without involving the spatial planning zone variable, which is Rp. 205.718/m2. In other words, the results increase the level of accuracy by 8% from the model without spatial planning zones with an estimated average error of Rp. 225,262/m2 for the combination of total sample points and outlier-free test points in 2007. These results provide information that in 2007, the condition of land prices in the Eastern Bandung City is quite varied with some high residual values located in strategic areas, both from heterogeneous public facilities and from the spatial concept that has been regulated in the regulations. Thus, the spatial planning zone variable can be a new perspective in making a GWR-based land price estimation model in addition to the physical object variable in the form of public or social facilities, especially to improve the quality of the model formed.