Next Article in Journal
Prediction and Uncertainty Capabilities of Quantile Regression Forests in Estimating Spatial Distribution of Soil Organic Matter
Previous Article in Journal
A Method to Identify Urban Fringe Area Based on the Industry Density of POI
 
 
Article
Peer-Review Record

Simulating the Spatial Heterogeneity of Housing Prices in Wuhan, China, by Regionally Geographically Weighted Regression

ISPRS Int. J. Geo-Inf. 2022, 11(2), 129; https://doi.org/10.3390/ijgi11020129
by Zengzheng Wang 1, Yangyang Zhao 2,* and Fuhao Zhang 1,2
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Reviewer 3: Anonymous
Reviewer 4: Anonymous
ISPRS Int. J. Geo-Inf. 2022, 11(2), 129; https://doi.org/10.3390/ijgi11020129
Submission received: 12 December 2021 / Revised: 29 January 2022 / Accepted: 9 February 2022 / Published: 11 February 2022

Round 1

Reviewer 1 Report

This paper proposes  a regionally geographically weighted regression (RGWR) to improve the accuracy of GWR estimation by taking an example of the residential sales price in Wuhan City. 
Some suggestions for the development of the paper are as follows.

This paper needs further explanation of why the analysis is based on the relationship between commercial house price and location and other factors.
In particular, although the subject of analysis is commercial house price, there are few literature studies related to house price due to the focus on methodology, so it needs to be supplemented.
Also, there are various studies on GWR. It is necessary to mention the distinction from other studies.

line 57-59: . affected -> delete . Sentences need to be clarified.
line 61-64: Provide separate sources for the fact that the house price is £18,000 higher and that it is 8% higher.
line 75: it the concept of "region" is not considered -> articulate the sentence.
line 64-67: Please add sentences regarding the reason.
The spatial autoregression issue is well explained, but a more detailed explanation is needed on why a dummy variable is used.
The distinction and difference between this study and previous studies need to be further emphasized.
line 97-101: u_i,v_i,β_k, β_i please use appropirate subscript.
More comments on why the authors should use this method are needed.
line 166-168 "Where b is called bandwidth, if point i and point j in the same region, which is standard geographically weighted regression. If point i and point j in the same region, which 
Wij=0 , is regionally and geographically weighted regression." need to be corrected.
line 234: "Use administrative region to constrain the calculation range of geographically weighted regression." please complete the sentence.
line 284-285: what do 0.48w, 5.38w, and 1.85w stand for?
line 287 "Jiang'an District. Court. " is this correct?
In relation to Table 4, it is not properly indicated whether the relationship between coefficient and house price is statistically significant, so it needs to be supplemented.
In the conclusion and analysis results (line 431-448), it would be good to add the comparison of results with other literature.

Author Response

请参阅附件。

Author Response File: Author Response.pdf

Reviewer 2 Report

Ιn this paper, the authors propose a regionally geographically weighted regression (RGWR). The residential sales price in Wuhan City is taken as an example to analyze from three aspects: model performance, fitting effect and influencing factors.
Starting with a general review, I have to point out that the paper is well-organized and has the proper structure. It is not scientifically thorough in the description and is not well-proof-read. The number of references is satisfactory but, many of them are not up-to-date. Based on the performance metrics proposed by the authors, the proposed method seems to have better behaviour.
Specific issues that need further explanation and improvement are the following.
1)The authors omitted the section"Related Work". What are the main challenges in this domain? What are the limitations of the previous works, which motivate the current study? How your study is different from others? What’s new/novel here?
2)The authors make a brief description in section 2, "Materials and Methods" about Geographical weight regression and Regionally and geographically weighted regression.
The article lacks problem formulation. There is no specific methodology and, it is not presented step by step.
Ηοw do the authors exploit the GWR and RGWR models?
3)A serious limitation is that the available dataset contains 954 observations. That creates problems in the reliability of the results.
We can not draw definite conclusions about the accuracy of the models.
4)The "Experiment analysis" section lacks documentation.
Under what conditions did the experimental results appear? What parameters were set?
5)What future directions emerge from this article?

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Very interesting paper, well structured and developed with a clear definition of objectives, methodological design, and sound discussion of results.

Author Response

We would like to gratefully thank the reviewer for his/her constructive comments. We have extensively revised the entire article (including Introduction, Materials and Methods, Experiment analysis, Discussion, and Conclusions section), clarifying the questions of the existing research: existing research little attention has been paid to the concept that both spatial non-stationary features and spatial discrete heterogeneity exist. We explain local effects in space by proposing the RGWR model. Solve the problem of simultaneously detect the spatial non-stationary features and spatial discrete heterogeneity in housing prices, these findings could improve global spatial regression analysis research and theory. In addition, we proofread the full text for English expressions, and at the end we asked the language editor of MDPI to conduct a language check on the article.

 

We heartfelt gratitude reviewers for their comments and suggestions!

 

Reviewer 4 Report

The manuscript deals with the adjustment of Geographically weighted regression in simulating housing prices in a selected location.
Suggestions and comments:
Edit the explanation of abbreviations in the text - the abbreviation GWR (Row 13) is used in the abstract
Need to emphasize the aim of the paper by creating a new paragraph (Row 80)
Does the article address only the price of real estate (in the case of buying and selling) or does it also address the price of renting a property?
What was the mix of properties - single houses vs. flats?
Does the calculation also take into account the total cost of housing (electricity, heating bills)?

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors have sufficiently improved the manuscript to warrant publication in IJGI. 

Reviewer 2 Report

I have no additional remarks on the revised version.

Reviewer 4 Report

Thank you for drafting the answers to my questions and revising and completing the manuscript. The article is also interesting to me for the comparison of differences in approach to housing issues.

Back to TopTop