Applying Hybrid ARIMA-SGARCH in Algorithmic Investment Strategies on S&P500 Index
Round 1
Reviewer 1 Report
Review of the Manuscript Entropy-1496574 „Applying Hybrid ARIMA-SGARCH in Algorithmic Investment Strategies on S&P500 Index” for the Entropy Journal.
General Comments
From my point of view, it is a very interesting topic and simultaneously it seems that to the best of my knowledge is the first empirical research aims to compare the performance of ARIMA as a linear model with that of the combination of ARIMA and GARCH family models to forecast S&P500 log returns in order to construct algorithmic investment strategies on this index. The paper consists of the following sections: Introduction, Literature Review, Data Analysis, Methodology, Empirical Results and Discussion, Robustness Test and Conclusions.
However, I find some recommendations:
- We recommend that the paper be structured in 4 parts: Introduction, Literature review, Method and Results, and Conclusions.
- The abstract must contain the main purpose of the paper, the research method used in the research and the main contributions.
- It would be very useful to add in the "Introduction" section the purpose, objectives and hypothesis of the research.
- We recommend that the authors present the opinions of specialists in a different section that can be called “Literature review”.
- We consider that the introduction should specify the novelty of the paper compared to other papers published in this area.
- Also, we consider the literature is not enough and that is why, we recommend the authors to refer to other recent works indexed in Web of Science, Scopus, Emerald, Cambrige, and of course MDPI Journals. We suggest that the authors cite papers published in MDPI journals and Web of Science Journals, such as:
- Nichita, A., Batrancea, l., Pop, C.M., Batrancea,I., Morar, I.D., Masca, E., Cesar, A.M.R., Forte, D., Formigoni, H., da Silva, A.A. (2019). We learn not for school but for life: Empirical evidence of the impact of tax literacy on tax compliance. Eastern European Economics 57: 397–429.
- Batrancea, L. (2021). An Econometric Approach Regarding the Impact of Fiscal Pressure on Equilibrium: Evidence from Electricity, Gas and Oil Companies Listed on the New York Stock Exchange, Mathematics9, no. 6: 630. https://doi.org/10.3390/math9060630.
7. In the econometric section the authors apply the fixed effect estimation or the random effect estimation (see for instance, Baltagi (2008), Hsiao (2014) and AndreB et al. (2015)). Besides, the corresponding tests to determine which is the best method of estimation is needed (see the Hausman test, the Breusch and Pagan (1980)´s Lagrange multiplier, the F test for fixed effects to test whether all unobservable individual effects are zero).
8. Additionally, they should test whether there is endogeneity because this problem must be taking into account in the estimation methodology.
9. The authors talk about the long-run relationship between these variables, however they do not support the empirical evidence providing panel cointegration tests that are crucial (see for instance Kao (1999) panel data cointegration test, the Pedroni (1999, 2004) panel data cointegration test or the Westerlund (2005) panel data cointegration test, among others).
10. It should be interesting that the authors estimate the marginal effects before and after the financial crises to compare before and after this particular structural break.
10. I think that descriptive statistics and correlation matrix is crucial to understand the context of the problem to be analyzed.
In conclusion, the article should be improve. It should also be enhanced with a review of the literature adequate to the subject and a broader interpretation and commentary of the research results.
The results of the research presented in the paper are valuable and should be presented to the scientific audience in the economic theory and practice. The paper can be published in Entropy Journal after minor revisions.
Author Response
Attached please find our cover letter with detailed responses to the Reviewer's comments.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Dear authors,
I express my conclusions about this article proposal through the following points of view
1) from 45-46 - in a scientific article there is no need to specify certain theoretical elements, definitions that are classical. Readers of this article should know what homoskedastic means.
2) Line 205
"The first step in the process of cleaning the data is to delete all missing and invalid"
For the missing data in the literature and in econometric practice we have the tool called imputation: by average, by regression, by machine learning.
Deleting data from my point of view leads to loss of knowledge about economic and financial phenomena.
3) In general for each of your models you have highlighted the statistical and econometric significance of the models. I consider that an economic-financial interpretation of the statistical meanings is more adequate.
4) In subparagraph 4.5.2 the authors present elements from the classical theory of statistics through equations 22-23-24-25. These theoretical elements are useful if we bring new, "theoretical" elements in econometrics, for example a new distribution function, a new metric, a statistical test different from those known so far.
In the case of articles dealing with applied elements of classical theories, it is not necessary to explain the fundamentals of statistics and econometrics.
5) Please extend the visibility of figures 7 and 8.
6) Please extend the bibliography with at least 10-15 bibliographic resources.
7) I have attached the similarity ratio, but I consider that the index of 17% is within a reasonable confidence interval.
Respectfully
Comments for author File: Comments.pdf
Author Response
Attached please find our cover letter with detailed responses to the Reviewer's comments.
Author Response File: Author Response.pdf
