Research on a Prediction Model and Influencing Factors of Cross-Regional Price Differences of Rebar Spot Based on Long Short-Term Memory Network

Round 1

Reviewer 1 Report

The authors address an important issue in their study. However, there are some issues to be addressed.

Abstract:

Needs to be improved, including some discussion and conclusions.

Introduction:

The introduction should be summarized.

There are several statements that need to be referenced. The authors only include 5 references to support their study.

Which tools were used before to predict steel prices, and what are their limitations?

Why the “long short-term memory network” is a better tool to predict steel prices than other tools?

Literature review.

The authors need to deepen on the discussion of the tools used to predict the price of metals in general and steel rebars. Furthermore, they need to better highlight the benefits of considering the cross-regional price difference and spot commodity price discovery at the same time.

Data preparation.

The authors mention that “Some indicators in this study have data missing values” and that “Indicators with a large proportion of missing values are removed” how this affects the projections?

Figure 1: Please indicate the nomenclature for y-axis and discuss the figure.

Sections 3, 4.1, and 4.2 should be rearranged in a “materials and methods section”, while section 4.3 should be an independent section of results.

A discussion section is needed to discuss the implications of the results obtained in this study and how companies can benefit from it.

Conclusions:

Conclusions need to be rewritten. It looks more like a summary than the presentation of the concluding remarks of the paper.

References:

 

Update the reference lists to include more recent references.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

I recognize that the originality of this paper could be evaluated.

In addition I would like the authors to reinforce the following points.

1) Why did the authors choose LSTM?  There are some other models that are useful in predictive models, such as Bi-LSTM, multivariate CNN-LSTM, etc.

2) I would like the authors to carefully describe the hypothesis you want to verify and the fact that the hypothesis was verified.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Thanks for the quick revision work. I think this paper deserves acceptance. I hope it will catch the eye of many researchers.