Review Reports

Round 1

Reviewer 1 Report

This paper addresses an area of property price analysis that has not been covered in previous research in the residential property modelling area.

A sound literature review demonstrates the issues involved and the work carried out to date on spatial autocorrelation with the gap in the area well documented

The approach to the research is sound and the explanation of the construction of the model is logical and appropriate.

With a well defined study area and a substantial sales transaction data base the inputs are sound and provide a strong basis for the developed model and the subsequent results.

The only addition that i would suggest is a comment on how these results would be viewed in relation to the Hong Kong market post 2009

Author Response

Reviewer 1

This paper addresses an area of property price analysis that has not been covered in previous research in the residential property modelling area.

A sound literature review demonstrates the issues involved and the work carried out to date on spatial autocorrelation with the gap in the area well documented

The approach to the research is sound and the explanation of the construction of the model is logical and appropriate.

With a well defined study area and a substantial sales transaction data base the inputs are sound and provide a strong basis for the developed model and the subsequent results.

The only addition that i would suggest is a comment on how these results would be viewed in relation to the Hong Kong market post 2009

A short discussion has been included in the Discussion section to explain why the results should remain intact even a different time period of the sample is examined.

 

Reviewer 2

This is an interesting paper considering factors affecting spatial autocorrelation in residential property prices on the example of Hong Kong, however there is a number of issues that I would suggest authors to consider amending:

- considering the fact that the aim and scope of LAND Journal is to publish high quality original articles on latest research dealing with various aspects of, amongs others, all aspects of land use and land cover change, land tenure/ownership/disposal rights, including disputes, geopolitical trends and land grabs, landscape and territorial planning, conservation and management, I would recommend the authors to extend the literature review on the current state of art (latest publications from international journals) justifying that the paper fits into the Journals’ scope.

 

We note the reviewers suggestion and have identified through some studies the relationship between our study and the scope of the journal. We also believe that our study is in scope with this (Land’s) special issue relating to "Property in the Space: Real Estate Spatial Analysis, Land Use, Urban-Rural Interactions, Management and Valuation", as we test spatial autocorrelation in property prices which relates to real estate spatial analysis and also valuation. We have also incorporated one additional paragraph to the manuscript to discuss the current state of the art in mass valuation, briefly covering topics such as artificial intelligence, machine learning etc.

 

- taking into account the journals requirement that research papers should ideally go beyond describing narrow case studies, I would recommend the authors to consider widening the scope of studies to other cases if possible. The authors could take advantage of the fact that they represent academic institutions located in totally different parts of the world,

 

We thank the reviewer for the suggestion. We concentrate on the Hong Kong market given its market structure and intensive land use and generally ‘apartment living’ as this forms the basis of our hypothesis to do with vertical S.A.. We have accordingly made a suggestion in the last section that further studies could be undertaken to explore the spatial dynamics and autocorrelation in house prices using property transaction data of different international housing markets to “re-evaluate” the statistical robustness of our study.

 

- the authors is many parts of the paper refer to property valuation models, either the traditional ones based on the comparative approach or automated solutions (e.g. mass valuation). I think that it would be helpful for potential readers to refer to current state of art considering the problems of property valuation and attempts of their solution by e.g. utilizing human emotion recognition in the significance assessment of property attributes or utilizing geoscience methods in real estate market analyses subjectivity decrease, just to underline how problematic property valuation can be. 

 

We thank for the suggestions made by the reviewer. A new paragraph has been incorporated into the text to elucidate some common problems of traditional hedonic-based valuation methods, including those arising from dimensionality, computational intensiveness, and degree of parsimony. We have further explained why spatial lag models are superior in practice in terms of computational simplicity.

 

The research methodology has been prepared in a very comprehensible way. I do not have any recommendations for amendments in that part. The verification of the results is also well prepared. In the conclusions I would recommend the authors to underline once again what the original solution proposed in the paper brings to the current state of art and why the proposed solutions outperforms the existing ones.

 

We have modified the conclusions according to the reviewer’s comment.

 

Reviewer 3

No specific comments.

Reviewer 2 Report

This is an interesting paper considering factors affecting spatial autocorrelation in residential property prices on the example of Hong Kong, however there is a number of issues that I would suggest authors to consider amending:

- considering the fact that the aim and scope of LAND Journal is to publish high quality original articles on latest research dealing with various aspects of, amongs others, all aspects of land use and land cover change, land tenure/ownership/disposal rights, including disputes, geopolitical trends and land grabs, landscape and territorial planning, conservation and management, I would recommend the authors to extend the literature review on the current state of art (latest publications from international journals) justifying that the paper fits into the Journals’ scope.

- taking into account the journals requirement that research papers should ideally go beyond describing narrow case studies, I would recommend the authors to consider widening the scope of studies to other cases if possible. The authors could take advantage of the fact that they represent academic institutions located in totally different parts of the world,

- the authors is many parts of the paper refer to property valuation models, either the traditional ones based on the comparative approach or automated solutions (e.g. mass valuation). I think that it would be helpful for potential readers to refer to current state of art considering the problems of property valuation and attempts of their solution by e.g. utilizing human emotion recognition in the significance assessment of property attributes or utilizing geoscience methods in real estate market analyses subjectivity decrease, just to underline how problematic property valuation can be. 

 

The research methodology has been prepared in a very comprehensible way. I do not have any recommendations for amendments in that part. The verification of the results is also well prepared. In the conclusions I would recommend the authors to underline once again what the original solution proposed in the paper brings to the current state of art and why the proposed solutions outperforms the existing ones.

Author Response

Reviewer 1

This paper addresses an area of property price analysis that has not been covered in previous research in the residential property modelling area.

A sound literature review demonstrates the issues involved and the work carried out to date on spatial autocorrelation with the gap in the area well documented

The approach to the research is sound and the explanation of the construction of the model is logical and appropriate.

With a well defined study area and a substantial sales transaction data base the inputs are sound and provide a strong basis for the developed model and the subsequent results.

The only addition that i would suggest is a comment on how these results would be viewed in relation to the Hong Kong market post 2009

A short discussion has been included in the Discussion section to explain why the results should remain intact even a different time period of the sample is examined.

 

Reviewer 2

This is an interesting paper considering factors affecting spatial autocorrelation in residential property prices on the example of Hong Kong, however there is a number of issues that I would suggest authors to consider amending:

- considering the fact that the aim and scope of LAND Journal is to publish high quality original articles on latest research dealing with various aspects of, amongs others, all aspects of land use and land cover change, land tenure/ownership/disposal rights, including disputes, geopolitical trends and land grabs, landscape and territorial planning, conservation and management, I would recommend the authors to extend the literature review on the current state of art (latest publications from international journals) justifying that the paper fits into the Journals’ scope.

 

We note the reviewers suggestion and have identified through some studies the relationship between our study and the scope of the journal. We also believe that our study is in scope with this (Land’s) special issue relating to "Property in the Space: Real Estate Spatial Analysis, Land Use, Urban-Rural Interactions, Management and Valuation", as we test spatial autocorrelation in property prices which relates to real estate spatial analysis and also valuation. We have also incorporated one additional paragraph to the manuscript to discuss the current state of the art in mass valuation, briefly covering topics such as artificial intelligence, machine learning etc.

 

- taking into account the journals requirement that research papers should ideally go beyond describing narrow case studies, I would recommend the authors to consider widening the scope of studies to other cases if possible. The authors could take advantage of the fact that they represent academic institutions located in totally different parts of the world,

 

We thank the reviewer for the suggestion. We concentrate on the Hong Kong market given its market structure and intensive land use and generally ‘apartment living’ as this forms the basis of our hypothesis to do with vertical S.A.. We have accordingly made a suggestion in the last section that further studies could be undertaken to explore the spatial dynamics and autocorrelation in house prices using property transaction data of different international housing markets to “re-evaluate” the statistical robustness of our study.

 

- the authors is many parts of the paper refer to property valuation models, either the traditional ones based on the comparative approach or automated solutions (e.g. mass valuation). I think that it would be helpful for potential readers to refer to current state of art considering the problems of property valuation and attempts of their solution by e.g. utilizing human emotion recognition in the significance assessment of property attributes or utilizing geoscience methods in real estate market analyses subjectivity decrease, just to underline how problematic property valuation can be. 

 

We thank for the suggestions made by the reviewer. A new paragraph has been incorporated into the text to elucidate some common problems of traditional hedonic-based valuation methods, including those arising from dimensionality, computational intensiveness, and degree of parsimony. We have further explained why spatial lag models are superior in practice in terms of computational simplicity.

 

The research methodology has been prepared in a very comprehensible way. I do not have any recommendations for amendments in that part. The verification of the results is also well prepared. In the conclusions I would recommend the authors to underline once again what the original solution proposed in the paper brings to the current state of art and why the proposed solutions outperforms the existing ones.

 

We have modified the conclusions according to the reviewer’s comment.

 

Reviewer 3

No specific comments.

Reviewer 3 Report

It is a nice paper

Author Response

Reviewer 1

This paper addresses an area of property price analysis that has not been covered in previous research in the residential property modelling area.

A sound literature review demonstrates the issues involved and the work carried out to date on spatial autocorrelation with the gap in the area well documented

The approach to the research is sound and the explanation of the construction of the model is logical and appropriate.

With a well defined study area and a substantial sales transaction data base the inputs are sound and provide a strong basis for the developed model and the subsequent results.

The only addition that i would suggest is a comment on how these results would be viewed in relation to the Hong Kong market post 2009

A short discussion has been included in the Discussion section to explain why the results should remain intact even a different time period of the sample is examined.

 

Reviewer 2

This is an interesting paper considering factors affecting spatial autocorrelation in residential property prices on the example of Hong Kong, however there is a number of issues that I would suggest authors to consider amending:

- considering the fact that the aim and scope of LAND Journal is to publish high quality original articles on latest research dealing with various aspects of, amongs others, all aspects of land use and land cover change, land tenure/ownership/disposal rights, including disputes, geopolitical trends and land grabs, landscape and territorial planning, conservation and management, I would recommend the authors to extend the literature review on the current state of art (latest publications from international journals) justifying that the paper fits into the Journals’ scope.

 

We note the reviewers suggestion and have identified through some studies the relationship between our study and the scope of the journal. We also believe that our study is in scope with this (Land’s) special issue relating to "Property in the Space: Real Estate Spatial Analysis, Land Use, Urban-Rural Interactions, Management and Valuation", as we test spatial autocorrelation in property prices which relates to real estate spatial analysis and also valuation. We have also incorporated one additional paragraph to the manuscript to discuss the current state of the art in mass valuation, briefly covering topics such as artificial intelligence, machine learning etc.

 

- taking into account the journals requirement that research papers should ideally go beyond describing narrow case studies, I would recommend the authors to consider widening the scope of studies to other cases if possible. The authors could take advantage of the fact that they represent academic institutions located in totally different parts of the world,

 

We thank the reviewer for the suggestion. We concentrate on the Hong Kong market given its market structure and intensive land use and generally ‘apartment living’ as this forms the basis of our hypothesis to do with vertical S.A.. We have accordingly made a suggestion in the last section that further studies could be undertaken to explore the spatial dynamics and autocorrelation in house prices using property transaction data of different international housing markets to “re-evaluate” the statistical robustness of our study.

 

- the authors is many parts of the paper refer to property valuation models, either the traditional ones based on the comparative approach or automated solutions (e.g. mass valuation). I think that it would be helpful for potential readers to refer to current state of art considering the problems of property valuation and attempts of their solution by e.g. utilizing human emotion recognition in the significance assessment of property attributes or utilizing geoscience methods in real estate market analyses subjectivity decrease, just to underline how problematic property valuation can be. 

 

We thank for the suggestions made by the reviewer. A new paragraph has been incorporated into the text to elucidate some common problems of traditional hedonic-based valuation methods, including those arising from dimensionality, computational intensiveness, and degree of parsimony. We have further explained why spatial lag models are superior in practice in terms of computational simplicity.

 

The research methodology has been prepared in a very comprehensible way. I do not have any recommendations for amendments in that part. The verification of the results is also well prepared. In the conclusions I would recommend the authors to underline once again what the original solution proposed in the paper brings to the current state of art and why the proposed solutions outperforms the existing ones.

 

We have modified the conclusions according to the reviewer’s comment.

 

Reviewer 3

No specific comments.

Reviewer 4 Report

This paper presents the results of a study of the determinants of spatial autocorrelation in residential property prices using the Hong Kong residential market as an example. In spite of advanced statistics, the results are somewhat controversial, which is due to some assumptions of geo-statistical methods in modeling the real estate market that are not entirely true, including the assumption that the closer properties are to each other the more similar (comparable) they are. This is not true due to the enormous variation of properties in geographic space in terms of function, use, structure, building type, etc.  Besides, these methods reduce the complex process of price formation by market participants to the evaluation of location (spatial) advantages of the property, while there are plenty of other factors and motives that people take into account when setting the transaction prices, and not always the spatial motives are predominant. The price of real estate is first born in the minds of market participants (not some real estate traders), so it depends primarily on social perception of many attributes of real estate, including spatial factors (not expressed in geographical coordinates), then the parties negotiate and come to a consensus between them, as a result of which the price level is determined.  This means that leaving aside social factors (buyers' and sellers' motives), which are locally, regionally and nationally differentiated, one cannot satisfactorily explain property prices (even if one processes thousands of geo-referenced data). Because of the enormous specificity of local housing markets, it is therefore difficult to rationally pursue the construction of global models.  The controversial conclusion that market liquidity (as total market trading volume) tends to increase both vertical and horizontal S.A., while market volatility (as standard deviation of the housing returns) is more prone to increase vertical S.A. but depress horizontal S.A. requires further explanation and justification in this context. It is generally assumed that it is not market liquidity or volatility, but rather the dynamic relationship between supply and demand in the market that will be fundamental to the phenomenon of spatial autocorrelation. Please comment on this problem in this context.

Specific comments:

1) In the title, abstract and conclusion, it should be made more clear that the research is on a specific housing market in Hong Kong, as otherwise the title suggests that the findings are applicable to all residential property markets, which is not the case, as housing markets are eminently local in nature and it is futile to look for universal regularities.

2) The authors use the term "apartment" (housing unit) interchangeably with the term "house", which is unacceptable, not only when talking about the so-called vertical spatial autocorrelation, but mainly because these are different segments of the housing market.  This should be corrected.

4) The authors also use the concept of "price" and "value" of residential real estate quite freely, and these are different categories and cannot be equated with each other. This should be corrected and definitions of terms used should be provided. 

5) Please also explain why the authors present 2022 research on historical data from 1998-2009. Land is not a historical journal after all.

Despite the criticisms, I think the paper is valuable because of new elements (such as the concept of 3D spatial autocorrelation analysis) and may be of interest to an international academic reader. Hence, after taking into account the above remarks and making revisions, it may be published in the scientific Land journal.

Author Response

Reviewer 1

This paper addresses an area of property price analysis that has not been covered in previous research in the residential property modelling area.

A sound literature review demonstrates the issues involved and the work carried out to date on spatial autocorrelation with the gap in the area well documented

The approach to the research is sound and the explanation of the construction of the model is logical and appropriate.

With a well defined study area and a substantial sales transaction data base the inputs are sound and provide a strong basis for the developed model and the subsequent results.

The only addition that i would suggest is a comment on how these results would be viewed in relation to the Hong Kong market post 2009

A short discussion has been included in the Discussion section to explain why the results should remain intact even a different time period of the sample is examined.

 

Reviewer 2

This is an interesting paper considering factors affecting spatial autocorrelation in residential property prices on the example of Hong Kong, however there is a number of issues that I would suggest authors to consider amending:

- considering the fact that the aim and scope of LAND Journal is to publish high quality original articles on latest research dealing with various aspects of, amongs others, all aspects of land use and land cover change, land tenure/ownership/disposal rights, including disputes, geopolitical trends and land grabs, landscape and territorial planning, conservation and management, I would recommend the authors to extend the literature review on the current state of art (latest publications from international journals) justifying that the paper fits into the Journals’ scope.

 

We note the reviewers suggestion and have identified through some studies the relationship between our study and the scope of the journal. We also believe that our study is in scope with this (Land’s) special issue relating to "Property in the Space: Real Estate Spatial Analysis, Land Use, Urban-Rural Interactions, Management and Valuation", as we test spatial autocorrelation in property prices which relates to real estate spatial analysis and also valuation. We have also incorporated one additional paragraph to the manuscript to discuss the current state of the art in mass valuation, briefly covering topics such as artificial intelligence, machine learning etc.

 

- taking into account the journals requirement that research papers should ideally go beyond describing narrow case studies, I would recommend the authors to consider widening the scope of studies to other cases if possible. The authors could take advantage of the fact that they represent academic institutions located in totally different parts of the world,

 

We thank the reviewer for the suggestion. We concentrate on the Hong Kong market given its market structure and intensive land use and generally ‘apartment living’ as this forms the basis of our hypothesis to do with vertical S.A.. We have accordingly made a suggestion in the last section that further studies could be undertaken to explore the spatial dynamics and autocorrelation in house prices using property transaction data of different international housing markets to “re-evaluate” the statistical robustness of our study.

 

- the authors is many parts of the paper refer to property valuation models, either the traditional ones based on the comparative approach or automated solutions (e.g. mass valuation). I think that it would be helpful for potential readers to refer to current state of art considering the problems of property valuation and attempts of their solution by e.g. utilizing human emotion recognition in the significance assessment of property attributes or utilizing geoscience methods in real estate market analyses subjectivity decrease, just to underline how problematic property valuation can be. 

 

We thank for the suggestions made by the reviewer. A new paragraph has been incorporated into the text to elucidate some common problems of traditional hedonic-based valuation methods, including those arising from dimensionality, computational intensiveness, and degree of parsimony. We have further explained why spatial lag models are superior in practice in terms of computational simplicity.

 

The research methodology has been prepared in a very comprehensible way. I do not have any recommendations for amendments in that part. The verification of the results is also well prepared. In the conclusions I would recommend the authors to underline once again what the original solution proposed in the paper brings to the current state of art and why the proposed solutions outperforms the existing ones.

 

We have modified the conclusions according to the reviewer’s comment.

 

Reviewer 3

No specific comments.

Round 2

Reviewer 2 Report

The quality of the paper has been improved. The authors according to the suggestions:

·    reedited the literature review part so that it corresponds to the Journal’s scope,

·    referred to the current state of art so that the scientific problem has been underlined,

·    provided proper justification for the method proposal,

Concluding the above, the authors amended the paper according to suggestions. The quality of the paper has been improved. I think that the readers will appreciate the paper in the present form. I recommend the paper for publication in that form and would like to congratulate the authors for interesting original paper preparation.