Do Futures Prices Help Forecast Spot Prices? Evidence from China’s New Live Hog Futures
Round 1
Reviewer 1 Report
11. I like table 1. Good idea to include it. However, there really isn’t a deliverable grade for the CME contract as it is a cash settled contract thus the contract essentially matches the aggregate hog grade included in the lean hog index used to settle the contract.
22. More explanation of what the cash hog price is, how it is collected, how it is calculated etc. in the Wind database is needed.
33. Table 4. Statistical significance of differences in each accuracy measure should be noted across 1-7 in each row. Same for tables 6 & 7.
44. The very short time period for forecast evaluation is of course a major weakness that should be acknowledged.
55. You have continuously overlapping forecast horizons implying one forecast to the next in order likely have high error correlations. It would be interesting to demonstrate the tendencies of the various techniques to be biased one way or the other for a while in out-of-sample forecasting. The bias and associated time-varying dimension of bias is another useful forecast performance measure.
none
Author Response
Dear Reviewer,
The authors sincerely appreciate anonymous reviewer’ suggestions and comments. Those invaluable comments were helpful for improving the quality of our manuscript. This revision seriously considers and carefully addresses all suggestions and comments.
Below please find detailed responses to reviewer’ comments. Note that references cited in this response are attached at the end.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Do Futures Prices Help Forecast Spot Prices? Evidence from China’s New Live Hog Futures
This study assesses the forecasting performance of the live hog futures listed on the Dalian Commodity Exchange on future hog spot prices. The authors build six futures-based forecasting models from January 2021 to March 2023. The results show that China’s live hog futures prices help forecast forthcoming hog spot prices. In particular, the considered models tend to produce more accurate forecasts than the no-change model across different prediction horizons. Futures spread-based models generally have the best forecasts for short-run forecasting, while the simple linear regression is the best for medium-term forecasting.
The study investigates an interesting theme that deserves attention. However, the contributions to the literature, more discussion, and the importance of the findings should be strongly improved. Specific comments are given in what follows.
Major Comments
Introduction
· Overall, the introduction needs to improve. The authors should better motivate their study by explaining why it is necessary, and what policy-level problem is addressing. The authors should elaborate on the contributions to the existing literature and should evidence the gap in research.
· Line 23. Add the source after the first sentence “…China’s consumer price index”.
· Line 26. Add the source “…had risen in 2021.”
· Line 38. Add a more recent study Algieri and Kalkuhl, 2019 before Carter & Mohapatra, 2008.
· A part of the literature linked to financialization of commodity markets should be also mentioned. The livestock sector has four commodities: feeder cattle, lean hogs, live cattle, and pork bellies. Indeed, increases in measures of futures market trading volume might have led to increases in spot price growth and volatility. In particular, the works of Tang and Xiong (2012); Cheng and Xiong (2014); Algieri et al. (2017) should be considered.
Background
- Line 121. Add the source after the first sentence “…Chinese households”.
- Line 147. Specify the source of the data.
- Line 148. Specify if there is a first generic futures contract for live hogs, as generally presented in databases such as Bloomberg and Eikon.
- Line 197. More info should be provided on the calculation of futures, it is not sufficient SPAN system.
- Provide graphs about spot and futures of live hogs.
Experimental Setting
· Line 314. Explain the reason of having a window of 100 observations,
· Line 364. The authors state “ seven model examined”, please be clear are six, or specify in more details.
· Line 387. The authors should clarify the reason why in Table 5, they start counting from (2) to (7).
· The authors should give more details about the no-change model and its economic intuitions.
Economic discussion
· More elaboration on the economic implications of the findings and why are the results sensitive to the type of futures considered.
· The authors should discuss about the efficiency of live hogs futures markets and whether futures prices are unbiased predictors of future spot prices as the market efficiency theory postulates.
Conclusion
· Line 405. Avoid and leave out the reference to African swine fever and explain the reason of hog price volatility.
· The Policy implications should be deepened.
References:
Algieri B., Kalkuhl M. (2019). Efficiency and Forecast Performance of Commodity Futures Markets, American Journal of Economics and Business Administration, 11(1): 19-34.
Tang K., Xiong W. (2012) Index Investment and the Financialization of Commodities, Financial Analysts Journal, 68(6): 54-74.
Cheng I.H., Xiong W. (2014) The financialization of commodity markets, Annual Review of Financial Economics, 6(1): 419- 441.
Algieri B., Kalkuhl M., Koch N. (2017). A tale of two tails: Explaining extreme events in financialized agricultural markets, Food Policy, 69: 256-269.
Minor editing of English language required
Author Response
Dear Reviewer,
The authors sincerely appreciate anonymous reviewer’ suggestions and comments. Those invaluable comments were helpful for improving the quality of our manuscript. This revision seriously considers and carefully addresses all suggestions and comments.
Below please find detailed responses to reviewer’ comments. Note that references cited in this response are attached at the end.
Author Response File: 
Author Response.pdf
Reviewer 3 Report
There is however one comment and suggestion I would make. It is the following:
The paper does a good job at evaluating statistical performance of the forecasting models considered but provides no indication on how useful these forecasts can be in practice from the point of view of a market participant who might want to use them as guidance for trading. This is also important from the point of view of the implications of the findings for assessing market efficiency. so I would suggest reporting for the various models not just the RMSE, MAPE and SMAPE, which are useful to assess the forecasting models on statistical grounds, but also the in-sample and out-of-sample R^2. Please see Potì (2018) [Valerio Potì, 2018. "A new tight and general bound on return predictability," Economics Letters, Elsevier, vol. 162(C), pages 140-145.] on how this measure of forecasting performance can be used to judge whether the forecasts can be used to make trading profits and for testing market efficiency. If this is too much work, the authors could even just mention it as an extension that they will address in future work.
Author Response
Dear Reviewer,
The authors sincerely appreciate anonymous reviewer’ suggestions and comments. Those invaluable comments were helpful for improving the quality of our manuscript. This revision seriously considers and carefully addresses all suggestions and comments.
Below please find detailed responses to reviewer’ comments. Note that references cited in this response are attached at the end.
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
The author has revised the paper along the suggested lines. The revision has helped the author to refine his contribution to the literature and to better reframe the empirical analysis. I have appreciated the effort, therefore I would now consider the manuscript for publication.
Minor editing of English language required
Author Response
Thank you. We have the paper thoroughly checked and revised by our professional language editor—Barbara Nordin (e-mail: [email protected]), who specializes in language editing.
