Price Forecasting and Span Commercialization Opportunities for Mexican Agricultural Products
Reviewer 1 Report
The authors have great intentions to produce forecasting methods to assist the farmers. But as always the lack of data is the very reason why forecasting is limited in scope and usefulness.
I expected to see the actual prices of the agricultural commodities included in the study. And also, the comparison of the actual and predicted prices during the period under study. That way, the authors can graphically convince the readers of the relative accuracy of the forecasting models.
Are there no significant trends that can be detected among related agricultural products that by knowing the trends in some products you can closely impute the trends in future prices of other products.
Is forecasting more on determining the range of prices or the direction of the changes in the future? If the prices are expected to go down, by how much can the farmers absorb the decline and still remain in business?
We want to thank you for your comments,
According to these:We added plots for 14-week predictions for both main examples: Hass Avocado and big Valencia Orange. They are some products related. For example, the price of big nopal is related to the price of several other products. We added a future work paragraph where VAR models might be an alternative to improve the predictions. However, VAR model is part of a future work. Most of SARIMA approaches are good to predict the direction of the variable. In this case, the models were good for both: trend and range of prices. The range of prices, however, is limited to a small window: 14 weeks approximately.
Moreover, the introduction was slightly improved and the little discussion about your concerns such as “If the prices are expected to go down, by how much can the farmers absorb the decline and still remain in business?” were included.
Reviewer 2 Report
Thank you for the opportunity to review this paper. I find the forecasting exercise in this manuscript to be very interesting, given its technical approach and its focus on the Mexican horticultural sector. I have several concerns, however, that I hope that the authors can easily address as they revise the paper. First, the paper does not provide a consistent reporting of their methodology's potential usefulness to farmers. The abstract indicates, "The results suggest the feasibility for the implementation of systems to provide information for better decisions by Mexican farmers." The conclusion, in contrast, is more pessimistic, pointing out that the methodology "has limitations due to a) incomplete and ambiguous data bases, b) the feasibility for decisions based on this approach." I find the study to be of interest either way, but the authors' work would benefit from offering a single interpretation of the commercial applications of their work.
Second, much of the Mexican horticultural sector is export-oriented, while SNIIM provides price data for Mexican wholesale markets. Readers would benefit from more information identifying and quantifying the portion of the horticultural market that is covered by the SNIIM price data, along with an assessment of what segment of the horticultural sector would benefit from the price analysis presented in this manuscript. I can imagine a situation in which the bulk of a particular commodity (say, Hass avocados) is destined for to be exported, as well as a situation where the prices are determined in advance by contracts between the buyer and the seller. In such instances, it is possible that the methodology is of limited usefulness.
Third, I find figures 2 and 3 to be extremely difficult to read and interpret. While figures of such great detail would be fascinating to inspect on a poster at an academic conference, some different form of visual presentation is needed for a journal article.
Finally, the quality of the writing would benefit from a little more attention. the first sentence of the abstract is circular: essentially, "decision-making helps decision-making." I would write instead something along the lines of "Using data analysis to predict future trends in the agricultural sector can help decision-making." There are some long compound sentences that could be divided into separate sentences, and throughout the paper, a professional editor might be able to offer some minor suggestions for its revision.
All in all, I enjoyed reading this manuscript and wish the authors success in its revision.
We want to thank you for your comments,
According to these:The conclusion was improved, we consider that in our second step involves the productive information of the farmers to calculate their future profits. The final goal is to develop an App to help the farmers in their planning for planting and harvesting in order to improve their profits. SNIIM covers, at least, one wholesale market evert state in the country. Due to wholesale markets works as hubs, SNIIM data represents a wide picture of Mexico’s market. SNIIM description was added in the discussion section to highlight its high reliability. We made some changes in the abstract to avoid circular sentences. About the figures, we changed their captions to help their interpretation.
Moreover, the introduction, discussion and, conclusion were slightly improved.