Econometric Analysis of SOFIX Index with GARCH Models
, Margarita Shopova 1, Tihomir Varbanov 1, Evgeni Ovchinnikov 1 and Angelin Lalev 2
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsDear Authors,
My review is attached in a separate PDF file.
Please read this review carefully and ensure all my comments are addressed in the new corrected version of this paper before its subsequent submission.
Kind regards,
Reviewer
Comments for author File: 
Comments.pdf
Author Response
Dear Reviewer, Please see the attachment.
We sincerely hope we have met your demands for correction!
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsAbstract - is weak. Expand.
Introduction - must be changed. Here the relevance of the topic, the importance of forecasting the SOFIX index should have been given. The literature given here - the interpretation of the research articles should be given in the literature review.
Materials and Methods - it's good.
Results -it's good.
Discussion - it's good.
Conclusions- is weak. Expand.
The Journal in which you want to publish the article and many other journals of the Mdpi have enough articles that match your article. Please provide quotes from articles published in these journals and relevant to your article.
Author Response
Abstract - is weak. Expand.
The abstract is expanded with a brief summary of the main contributions of the paper.
Introduction - must be changed. Here the relevance of the topic, the importance of forecasting the SOFIX index should have been given. The literature given here - the interpretation of the research articles should be given in the literature review.
The Introduction is separated from the Literature Review.
In the Introduction we added the redefined aim of the study. We enriched the importance of the study and the relevance of the topic. We highlighted the noveltis and the contributions of our research. Also, we stated a Research Hypothesis.
In the Literature Review section we added a paper review in order to enrich it in terms of application and comparison of GARCH models and machine learning approaches in analyzing equity indices. Also, we end the section with a summary of the main results from the review of the literature.
Materials and Methods - it's good.
Results -it's good.
Discussion - it's good.
Conclusions- is weak. Expand.
The Conclusion and the Discussions sections are united in one section.
The Journal in which you want to publish the article and many other journals of the Mdpi have enough articles that match your article. Please provide quotes from articles published in these journals and relevant to your article.
P.S. Please review the edited version of the manuscript in which major changes have been made!
Reviewer 3 Report
Comments and Suggestions for AuthorsIn "Forecasting Volatility of SOFIX Index with GARCH Models" the authors propose to investigate five different GARCH models (GARCH, EGARCH, IGARCH, Component GARCH (CGARCH) and GJR-GARCH) along with six distributions (Normal, Student's t, GED and their skewed forms), which are used to predict volatility for the Bulgarian stock index SOFIX.
In the Introduction they present their economic and financial reasons for carrying out this study and they do so in a well-structured, clear and convincing way. They also carry out a review of the literature on these subjects, mainly illustrative, not at all critical and in an inordinate quantity. It could well be summarized to the most relevant cases.
In section 2. Material and Methods, the models to be used are presented in a "school" way, in too much detail, which, as this text is aimed at scientists and not students, makes for tedious reading.
In 3.Results, the results are presented clearly and well-illustrated, with the usual defect of prolixity that sometimes seems to hide what is being said. It is necessary to opt for simpler, more direct language and save on words. In 4. Discussion this is better achieved. There is some improvement compared to the previous section.
Part 5. Conclusions is the best: short, simple, and assertive.
That said, it should be noted that this text does not add anything to knowledge in its scientific field. There is no innovation in the methodology or in the subject to be studied. As for prediction, one of the objectives of the study, the authors themselves are aware of the weakness associated with the use of simulations. In any case, predictions of more than 60 years in this field, with these or other models, are not credible.
Therefore, I cannot recommend publishing this text.
Author Response
In "Forecasting Volatility of SOFIX Index with GARCH Models" the authors propose to investigate five different GARCH models (GARCH, EGARCH, IGARCH, Component GARCH (CGARCH) and GJR-GARCH) along with six distributions (Normal, Student's t, GED and their skewed forms), which are used to predict volatility for the Bulgarian stock index SOFIX.
We changed the title and the aim of the research. The new title is “Econometric Analysis of SOFIX Index with GARCH Models”. The aim is changed to “the purpose of this paper is to model the daily returns of the oldest index of the Bulgarian Stock Exchange SOFIX since its launch using different variants of GARCH models".
In the Introduction they present their economic and financial reasons for carrying out this study and they do so in a well-structured, clear and convincing way. They also carry out a review of the literature on these subjects, mainly illustrative, not at all critical and in an inordinate quantity. It could well be summarized to the most relevant cases.
The Introduction is separated from the Literature Review.
In the Introduction we added the redefined aim of the study. We enriched the importance of the study and the relevance of the topic. We highlighted the noveltis and the contributions of our research. Also, we stated a Research Hypothesis.
In the Literature Review section we added a paper review in order to enrich it in terms of application and comparison of GARCH models and machine learning approaches in analyzing equity indices. Also, we end the section with a summary of the main results from the review of the literature.
In section 2. Material and Methods, the models to be used are presented in a "school" way, in too much detail, which, as this text is aimed at scientists and not students, makes for tedious reading.
We agree that in this section the models to be used are presented a bit in a "school" way, in too much detail. However, as long as the audience of the journal not necessarily consists of econometricians, we would prefer to give the methodological details here, rather than force the reader to search them elsewhere.
In 3.Results, the results are presented clearly and well-illustrated, with the usual defect of prolixity that sometimes seems to hide what is being said. It is necessary to opt for simpler, more direct language and save on words. In 4. Discussion this is better achieved. There is some improvement compared to the previous section.
The Results section is enriched in the following manner. First, the data is divided into training and test subsets. The estimation of forecast errors, based on the train and test data, further confirms the choice of the best ARMA(1,1)-CGARCH(1,1) model with Student’s t-distribution. A robustness test of the estimated models is included, too. In the simulation process two scenarios were added in respect to the starting point.
Part 5. Conclusions is the best: short, simple, and assertive.
The Discussion and the Conclusion sections are united.
That said, it should be noted that this text does not add anything to knowledge in its scientific field. There is no innovation in the methodology or in the subject to be studied. As for prediction, one of the objectives of the study, the authors themselves are aware of the weakness associated with the use of simulations. In any case, predictions of more than 60 years in this field, with these or other models, are not credible.
Therefore, I cannot recommend publishing this text.
P.S. Please review the edited version of the manuscript in which major changes have been made!
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsDear Authors,
Only part of my comments were incorporated in the text of the revised paper. Moreover, limiting the main aim of the paper, for forecasting to just econometric analysis highly reduced its value in terms of originality/novelty and significance of content.
Therefore, I suggest accepting it after minor revisions, i.e., after including other comments indicated in the following points for my initial revision:
5.a. -> The author should add a detailed description of which results were obtained during training and the testing period because currently, it is not clear. It should be indicated in each figure and table.
5..b. -> I would rather focus on forecasting
5.c. -> If the method is so sensitive to the starting point then maybe it is not proper for the main purpose of this research?
9. -> Please increase the number and quality of cited papers on similar topics
11.b. -> the same comment as for 5.a.
11.c. -> the same comment as for 5.a.
11.e. -> change in good direction but please add some other robustness check tests
11.f. -> So please describe ii accordingly because, from the description of Table 3 and the description above Table, one cannot obtain the information that level in the table means the logarithms of the daily return of the SOFIX index.
11.g. -> based on which periods these models were estimated and tested to use later in the simulation of the SOFIX index
Best regards,
Reviewer
Comments on the Quality of English Languagenone
Author Response
Dear Reviewer, Please see the attachment.
The answers to your questions and suggestions are coloured in blue.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsI have to agree that the authors made a notable effort to revolutionize their manuscript, resulting in a text whose objectives are more credible and interesting. Here are my congratulations,
Coherence, scientific consistency, and a greater awareness of the reality studied were gained.
In its present form it can be published as is.
Author Response
I have to agree that the authors made a notable effort to revolutionize their manuscript, resulting in a text whose objectives are more credible and interesting. Here are my congratulations,
Coherence, scientific consistency, and a greater awareness of the reality studied were gained.
In its present form it can be published as is.
Thank you very much for your positive feedback on the manuscript.
We would just like to point out that further improvements have been made in the present version, both in the introduction and the references, and in the description of the some results obtained.
