An Auxiliary Index for Reducing Brent Crude Investment Risk—Evaluating the Price Relationships between Brent Crude and Commodities

Round 1

Reviewer 1 Report

Dear Authors,

The revised version is satisfactory.

Author Response

Dear Reviewer,
I have submit a reply file. Thanks for your valuable comments and suggestions.
We have followed your valuable comments and suggestions to revise the manuscript.

Best Regards

Reviewer 2 Report

In this paper, the price relationships between crude oil (Brent) and 78 commodities is measured using a correlation test and regression model. The research results show that New York Harbor No. 2 heating oil spot price (FOB) can forecast the rise and fall of the monthly price of crude oil.

• The analysis is only a simple correlation test;
• The models have not a theoretical background. Just 78 commodities are believed to be determinants for the crude oil price, which is not right;
• The regression equations are not tested with specification error tests for robustness,
• The prices of many commodities are used as determinants of crude oil price, while crude oil price could be on the contrary the determinant of those commodities' prices;
• Data are not seasonally adjusted and are not tested for unit root. This fact makes all estimation results inappropriate.
• Half of the paper explains the conventional statistical and econometric terms (regression model, correlation, R squared, scatter plot etc.), which are well-known and do not add any academic value to this paper.

Author Response

Dear Reviewer,
I have submitted a reply file. Thanks for your valuable comments.
We have followed your comments to revise the manuscript.
Best Regards
Mu-Chun Hsiao 2021.4.13

Author Response File: Author Response.pdf

Reviewer 3 Report

After revision, article is certainly improved, mainly the purpose of the study is much clearer for the reader now. Inclusion of the part which tests the forecasts is certainly helpful.

My previous concerns were sufficiently adressed by the authors, mainly by clarification of the forecasting period. I would only caution authors not to use words "speculation" and "investment" carelessly. Regardless of the used method and its precision, short-term oil trading using proposed method(indicator) in the article is certainly the former.

Author Response

Dear Reviewer,
I have submitted a reply file. Thanks for your valuable comments and suggestions.
We have followed your valuable comments and suggestions to revise the manuscript.

Best Regards

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

My comments regarding the previous version of the paper are not addressed properly.

Author Response

Dear reviewer,

  We have submitted the reply file to you. The manuscript also has revised carefully as following your comments and suggestions. All tests also have operated. Moreover, the empirical results after seasonal adjustments are slightly better than that of no-seasonal adjustment. The success rate of this index increases from 82.98% to 84.62%. Thanks for your valuable suggestions.

Best Regards,

Author Response File: Author Response.pdf

Round 3

Reviewer 2 Report

The revised version has not big problems and it is acceptable.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Thank you for the opportunity to read interesting research.
In order to improve the manuscript, I have some comments:

1. There is no reference in the introduction - line 25-412, line 53-54.
2. The manuscript lacks references to the latest research in this field and comparison of results, e.g .:
   1. Roman&Gorecka&Domagała 2020: https://www.mdpi.com/1996-1073/13/24/6545
   2. Vo&Vu&Vo 2019: https://www.mdpi.com/1996-1073/12/7/1344
   3. Taghizadeh-Hesary&Rasoulinezhad&Yoshino 2019: https://www.sciencedirect.com/science/article/pii/S0301421518308486?via%3Dihub
3. The methods used are simple, it makes no sense to present formulas for Persona correlations, covariance, R2 coefficient, or formulas for the general form of the regression model. Only the methods used need to be mentioned and more attention should be paid to the presentation of the variables and the procedure.
4. It should be clearly stated what the research period was and justify the sub-periods in the tables.
5. The results are described in detail, but there is no broad discussion with other studies.
6. Line 581: you need to list these papers, please refer to the articles mentioned in point 2.

Reviewer 2 Report

Main problem of the study is its illogicality. First authors select wide range of various commodities, including agricultural crops and industrial metals, but they also include same commodity multiple times, namely crude oil (and its refined components). Citing the authors, pg. 9: "It is amazing to find out that the price change rates rxy of the four selected commodities, with the exception of crude oil, are above 0.82, reaching as high as 0.94." What is amazing about that, when the same commodity (or its refined components) traded in different places (WTI and Dubai crude) is also used for analysis? It is more or less stating the obvious and I doubt that it could be useful. Authors argue that (pg. 20): "This is significant useful for investors and analysts due to the reduction of risk in Brent Crude investment." Authors should demonstrate this in the paper, instead of explaining, for instance, simple correlation. Big profits were made in the trading units of oil majors during the last year's crisis (Q2) due to crude oil storage shortage, which led to negative WTI price. I suspect that findings presented in this study would be of no use in that situation and authors should prove otherwise in order to demonstrate that the article should be published. In conclusion, authors state: "After examining the price relationships of Brent Crude with 78 global commodities, some commodities show good price relationships with Brent Crude...". This should be restated not to "some" but "same", and then again, is this finding surpising and novel? Hardly so. 

Reviewer 3 Report

Applying Simple Linear Regression to monthly data for 1989.01 to 2019.10, the authors have examined the price relationships between crude oil (Brent) with 78 global commodities. The research findings show that the price of New York Harbor No. 2 Heating Oil Spot Price FOB can be used as a forecasting index for changes in the monthly price of crude oil.

 

The following problems:

• The definitions, explanations and equations from introduction to statistics such as correlation, covariance, R squared, scatter plot, OLS regression makes half of the paper. These definitions and explanations are not necessary. Only the calculation results should be used in the paper.
• “Procedure of This Study” given in the paper also is not necessary.
• Descriptive statistics are not appropriate.
• It is noticed that the data are real data and if the data are real data, how the real data is arranged?
• The data is monthly data. That is why it should be seasonally adjusted before using in calculations. The is not any word about the seasonal adjustment.
• The data is time-series data, but it is not tested for unit root. The presence of unit root affects the robustness of the estimations.
• Only simple regression is used without appropriate pre- and post-estimation tests for robustness.