An Analysis of the Price Determinants of Multiplex Houses through Spatial Regression Analysis

Round 1

Reviewer 1 Report

An Analysis on the Price Determinants of Multiplex Houses
through Spatial Regression Analysis
By

Jae-jong Kim, Mi-Jeong Cho, and Myeong-Hun Lee

Abstract: This study established a model for price determinants with the combination of GIS technique and spatial regression model based on the parcel prices of multiplex houses in an effort to integrate and utilize spatial data and chose a suitable model. This study established a spatial weights matrix to apply interrelation with adjacent areas and performed row standardization to specify the effect of adjacent areas. Moran’s I was used for measuring the spatial autocorrelation of the parcel prices of multiplex houses. Through this, the parcel price of multiplex houses was analyzed to have a strong spatial autocorrelation and be related with jeonse price of apartment. This study also found if the jeonse price of apartment would have an effect on the other housing types in the neighborhood. For creating a sustainable residential environment when redeveloping an aging residential area, there is a need to find various ways for coexistence by identifying the interrelation with the neighboring residential areas rather than simply focusing on the supply amount.

 

The introduction provides sufficient background and include all relevant references. The results are clearly presented.  The research design is appropriate. The methods are adequately described. The conclusions are supported by the results.

Minor revision:

1. Check and re-write the yellow part typos in the attached paper.
2. Paper must be write by using journal format. Checking the yellow part of the attached paper with minor revision, I recommend that the paper can be publish in the Sustainability.

Author Response

Dear reviewer, Thank you for giving me valuable comments to improve my article. I am not expert on this website the reason why I cannot find your attached file as following your comment. :) Even I have asked to MDPI information in order to find your file attached, but I couldn't get the reply back unfortunately. Please kindly understand and let me know where I could find your file attached with yellow color comment.   Q1) There is no attached file that i can find to see comments with yellow part.    Q2) I used resent journal format as I know. Please kindly let me know if there is certain journal format.    Thank you for your help.    Best regards,  Kim Jaejong.

Reviewer 2 Report

Overall comment

This study established a spatial weights matrix to apply interrelation with adjacent areas and performed row standardization to specify the effect of adjacent areas. The topic seems interesting and please kindly find below for my comments.

 

Detailed comments:

1. The research gaps and the research significance of the study need to be better presented or highlighted.
2. Section 3.3.2.1. As the construction of spatial weights matrix is one key step in this study, Rook contiguity approach is selected in this study without further explanation. A more rigorous examination of different types of spatial matrices (both distance-based and contiguity-based) are required to derive an ideal approach.
3. Section 3.1.2. The current view on spatial modelling selection process has extended beyond the framework proposed by Rosen (Figure 1). In order to select a suitable spatial regression, I would suggest the authors to look into the following paper written by Elhorst (2010) https://www.tandfonline.com/doi/full/10.1080/17421770903541772.
4. Insufficient literature review and references.

Author Response

Dear reviewer, 

Thank you for giving me valuable comments to improve my article.

I would like to share the result file attached and revised based on your comment with yellow color.

Please kindly see my revision version of the article.

In addition to this, the thing I would like to politely share is related with your detailed comment 3. I have completely read over and over again J. Paul Elhorst’s article as you have suggested. And here is my thought why I used this spatial modelling selection.

As a result of estimating spatial autocorrelation for multi-family housing prices in Seoul, it was found that spatial autocorrelation exists, making it inappropriate to estimate using the general lag model. select the model of Lagrange Multiplier Test is used to test the hypothesis about the spatial autocorrelation of the dependent variable and the error in the spatial quantitative model. Lagrange Multiplier Test is calculated as verification of the estimate of the error estimated through the maximum likelihood measurement. Compared to the OLS regression model, in order to determine which of the spatial disparity model and spatial error model is more effective and reliable than the OLS regression model, it is necessary to go through the process of testing as shown in the figure in the thesis submitted by the researcher.

If only one of the LM-Lag and LM-Error values ​​is statistically significant, the spatial regression model that appears to be significant can be selected. That is, when the LM-Lag value is significant, the spatial disparity model is applied, and when the LM-Error value is significant, the spatial error model can be used. As such, the criterion for determining which model is more suitable between the spatial disparity model and the spatial error model is LM-Lag. It is possible through the test result of LM-Error value. However, if both the LM-Lag and LM-Error values ​​are statistically significant, the statistical significance of the Robust LM-Error and Robust LM-Lag values ​​should be further tested, and the final spatial regression model should be selected according to the results.

In general, when the null hypothesis for spatial dependence is rejected, the spatial disparity model and the spatial error model are estimated respectively, and the results are compared and selected. The fit of the spatial regression model is tested with R2, log-likelihood, AIC, SC, etc. When the spatial regression model is used, the log-likelihood increases and AIC and SC decrease compared to the OLS regression model, improving the model's straightforwardness. Therefore, by comparing the log-likelihood values ​​of the OLS regression model with the log-likelihood values ​​of the spatial disparity model and the spatial error model, it can be seen that the model with more log-likelihood values ​​is more suitable. Similarly, by comparing AIC and SC values ​​in the same way, a spatial regression model with better fit is selected compared to the OLS regression model.

Hopefully, it would match and fully satisfy your standard and comment. And I look forward to getting your permission. ?

Thank you.

Best regards,

Kim Jaejong

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

The newly added references from [26] to [31] have not been found inside the text. Please add them accordingly.

Author Response

Dear reviewer,

Hope you are doing well.

It is very honored to receive your second comments, and I have attached revised version of file uploaded with yellow mark and memo of Microsoft Word.

And I would like to explain that I fixed all the reference numbers until 30th added 26th to 30th.

In addition to this, I deleted reference 31th since I have not written any of contents.

I hope I get your approval for this time.

Thank you.

Best regards,

Kim Jaejong

Author Response File: Author Response.docx