PCLM-D-26-00056

Rethinking Climate Econometrics: Data Cleaning, Flexible Trend Controls, and Predictive Validation

PLOS Climate

Dear Dr. Schötz,

Thank you for submitting your manuscript to PLOS Climate. After careful consideration, we feel that it has merit but does not fully meet PLOS Climate’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by May 13 2026 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at climate@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pclm/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

• A letter that responds to each point raised by the editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
• A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
• An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

As the corresponding author, your ORCID iD is verified in the submission system and will appear in the published article. PLOS supports the use of ORCID, and we encourage all coauthors to register for an ORCID iD and use it as well. Please encourage your coauthors to verify their ORCID iD within the submission system before final acceptance, as unverified ORCID iDs will not appear in the published article. Only  the individual author can complete the verification step; PLOS staff cannot  verify ORCID iDs on behalf of authors.

We look forward to receiving your revised manuscript.

Kind regards,

María Dolores Gadea Rivas

Academic Editor

PLOS Climate

Journal Requirements:

1. Please amend your detailed Financial Disclosure statement. This is published with the article. It must therefore be completed in full sentences and contain the exact wording you wish to be published.

a. State the initials, alongside each funding source, of each author to receive each grant.

b. State what role the funders took in the study. If the funders had no role in your study, please state: “The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.”

c. If any authors received a salary from any of your funders, please state which authors and which funders.

2. We ask that a manuscript source file is provided at Revision. Please upload your manuscript file as a .doc, .docx, .rtf or .tex.

If the reviewer comments include a recommendation to cite specific previously published works, please review and evaluate these publications to determine whether they are relevant and should be cited. There is no requirement to cite these works unless the editor has indicated otherwise.

Additional Editor Comments (if provided):

The reviewers agree that the manuscript addresses an important topic and that the empirical analysis is generally carefully implemented. However, they also raise a number of substantive concerns that need to be addressed before the paper can be considered for publication.

In particular, the manuscript would benefit from a clearer positioning within the broader climate econometrics literature, especially by engaging more explicitly with time series approaches and related issues such as non-stationarity, dynamics, and structural breaks. In addition, the motivation and interpretation of the proposed methodology should be sharpened, including a more precise use of econometric terminology and a clearer discussion of the contribution relative to existing methods. Reviewer 1 also suggests strengthening the empirical dimension of the paper, including a more explicit discussion of the data properties, potential dynamic specifications, and the economic insights derived from the analysis.

We therefore invite you to submit a major revision that carefully addresses all the points raised by both reviewers. We believe that, with a thorough revision along these lines, the paper could make a valuable contribution.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

-->Comments to the Author

1. Does this manuscript meet PLOS Climate’s publication criteria? Is the manuscript technically sound, and do the data support the conclusions? The manuscript must describe methodologically and ethically rigorous research with conclusions that are appropriately drawn based on the data presented.-->

Reviewer #1: Yes

Reviewer #2: Partly

**********

-->2. Has the statistical analysis been performed appropriately and rigorously?-->

Reviewer #1: Yes

Reviewer #2: Yes

**********

-->3. Have the authors made all data underlying the findings in their manuscript fully available (please refer to the Data Availability Statement at the start of the manuscript PDF file)?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception. The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.-->

Reviewer #1: Yes

Reviewer #2: Yes

**********

-->4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS Climate does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.-->

Reviewer #1: Yes

Reviewer #2: Yes

**********

-->5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)-->

Reviewer #1: The paper is interesting and well writen. The title is to broad for what the paper covers. They could mention explicitly in the title that it is an application of statistical learning to a panel of climate varibles to evaluate the impact on GDP. The paper provides detailed guidance on using this statistical approach with this climate data set and on aggregating climate information. Comments: 1. The dependent variable is stationary, I(0). What are the time series propertoes of the explanatory variables Z and its first difference? Are they trend-stationary of I(1), difference-stationary? That information will determine which transformation to use in the panel to avoid having an unbalance model. 2. Why should eqution (1) be static? It would much more infromative in climate terms to specify a dynamic panel with las dependent variables, etc. This can perfectly be done in machine learning approaches. See for example, the application of random forest ina dynamic context, Escribano and Wang (2021). 3. Could this nonparametric detrending be used to do out of sample forecast, which is one of the main advantages of using statistical learning methods? 5. Is any part of this statistical learning procedure an original theoretical contribution of this paper? I would suggest to reduced the statistical details of the paper that are known with reference to the corresponding literature and concentrate in the empirical climate application and what we have learned from it. 6. There is a literature on the heterogeneity of climate change that is important in this spaciotemporal approch and thta should be discuss. Desmet and Rossi-Hansberg (2024) and Gadea and Gonzalo (2026).

REFERENCES:

Desmet K. and E. Rossi-Hansberg(2024). "Climate Change Economics Over Time and Space". WP 32197 NBER.

Escribano and Wang (2021). “Mixed Random Forest, Cointegration, and Forescating Gasoline Prices”. International Journal of Forecasting (2021).

Gadea and Gonzalo (2026). "Regional heterogeneity and warming dominance in the United States".PLOS Climate.

Reviewer #2: This paper revisits an important part of the early climate econometrics literature and points out a number of weaknesses in empirical practices. The authors advocate a more data driven or machine learning oriented approach to the analysis. In particular, they argue that much of climate econometrics relies on fragile empirical procedures: influential panel regressions are sensitive to outliers, often use ad hoc time trend controls, do not fully address the range of dependence structures that may arise across countries and years, and rarely rely on formal model selection combined with genuine out of sample prediction for model selection. In response, the paper proposes a framework built around data cleaning, nonparametric trend controls, and predictive testing, among other practices.

I have several concerns in its current version.

First, I am concerned with the literature review. In fact, the paper reduces climate econometrics to panel regressions with fixed effects, ignoring, for instance, time series econometrics. Although this tradition may have been less prominent in the earlier stages of the field, it has become increasingly important. In particular, the monograph by Castle and Hendry (2020) offers an alternative perspective on climate econometrics; indeed, its title is precisely Climate Econometrics. By focusing almost exclusively on panel data methods and predictive performance, the authors overlook other key issues emphasized in that literature, such as non-stationarity, structural breaks, and the role of the underlying data generating process. As a result, the methodological scope of the paper is narrower than the title and general framing might suggest, and its conclusions may not fully reflect the broader set of tools and insights available in climate econometrics. Relatedly, some of the authors’ proposed ways of addressing issues such as non-stationarity may not be fully satisfactory from a time-series perspective, especially given that the data span more than 60 years at annual frequency.

Second, I do not think it is accurate to suggest that econometrics, more generally, does not pay attention to outliers. It may well be true that some contributions in climate econometrics have not treated them adequately, as can happen in many applied fields, but this should not be generalized to econometrics as a whole. For instance, the paper by Pretis et al. (2018) provides one example of econometric methods explicitly designed to deal with outliers and structural breaks that, of course, can be used in climate related applications as they show in one of their examples. See, also, Castle and Hendry (2020). Of course, I fully agree with the authors that failing to deal properly with outliers can be very harmful.

Page 8, lines 154–155: I find the sentence “Using differenced variables is sometimes called level regression as opposed to growth regression which uses the original variables” rather unfortunate from an econometric point of view. In econometrics, the terminology “levels” versus “differences” can also refer to the dependent variable. In particular, an analysis in growth rates or log-differences is typically described as an analysis in differences of the dependent variable, whereas an analysis with the dependent variable in levels is referred to as a levels regression. I understand that here the dependent variable is always a growth rate and that the authors are instead referring to whether the explanatory variables enter in levels or in differences. Still, in its current form the terminology is confusing. If the paper aims to engage econometricians and persuade them of its critique, the jargon should be made much clearer.

Regarding the computation of standard errors, I am sympathetic to the authors’ point that the assumptions often imposed on the error structure in panel data applications can be too strong. In practice, some empirical work in climate econometrics has indeed relied on restrictive covariance assumptions, and uncertainty is often not properly taken into account and discussed. In this respect, I think the paper makes a useful contribution.

The idea of using out of sample performance as a model selection criterion is, in principle, appealing and has long been present in econometrics. See, for example, García-Ferrer et al. (1987), where models are compared in terms of their out of sample one step ahead mean squared forecast errors. Thus, this is not a new idea in econometrics, and it is well accepted in the literature. However, its particular implementation here through 5-fold cross-validation in a time series setting is much less straightforward. This concern is not fully resolved by applying the procedure to “residualized” data: it may remove low-frequency components, but it does not in itself guarantee that the transformed series are free of dynamic dependence or that the resulting folds are informationally independent. This is particularly relevant if the right hand side of the GDP growth equation does not include lagged dependent variables, as in García-Ferrer et al. (1987).

More broadly, some of the problems identified by the authors may reflect omitted variables. If relevant variables are missing, the estimated effects attributed to included controls may be biased, unless orthogonality conditions between the included and not included regressors hold, and the residuals may display persistence for that reason.

More on this, from an economic perspective, it is difficult to accept that GDP growth rates are not affected by economic variables.

Finally, while the authors’ pseudo p-values is a potentially useful descriptive device for asking whether a model performs better than an uninformed random predictor, it is not equivalent to a formal test of forecast accuracy. The paper’s metric benchmarks each model against an artificial Gaussian random prediction rule with optimally chosen variance, so its interpretation is closer to “predictive content relative to a no-information null” than to “statistically superior forecasting performance.” More generally, throughout the manuscript terms such as test and significance, which have specific meanings in the statistics and econometrics literature, are sometimes used rather loosely. I think this should be avoided.

References

Castle, J., & Hendry, D. (2020). Climate Econometrics: An Overview. Foundations and Trends in Econometrics, 10, 145–322.

García-Ferrer, A., Highfield, R. A., Palm, F., & Zellner, A. (1987). Macroeconomic Forecasting Using Pooled International Data. Journal of Business & Economic Statistics, 5(1), 53–67.

Pretis, F., Reade, J. J., & Sucarrat, G. (2018). Automated General-to-Specific (GETS) Regression Modeling and Indicator Saturation for Outliers and Structural Breaks. Journal of Statistical Software, 86(3), 1–44.

**********

-->6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

Do you want your identity to be public for this peer review?  If you choose “no”, your identity will remain anonymous but your review may still be made public.

For information about this choice, including consent withdrawal, please see our Privacy Policy.-->

Reviewer #1: No

Reviewer #2: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

--> -->-->Figure Resubmissions:

-->-->While revising your submission, we strongly recommend that you use PLOS’s NAAS tool (https://ngplosjournals.pagemajik.ai/artanalysis) to test your figure files. NAAS can convert your figure files to the TIFF file type and meet basic requirements (such as print size, resolution), or provide you with a report on issues that do not meet our requirements and that NAAS cannot fix.-->-->

After uploading your figures to PLOS’s NAAS tool - https://ngplosjournals.pagemajik.ai/artanalysis, NAAS will process the files provided and display the results in the "Uploaded Files" section of the page as the processing is complete. If the uploaded figures meet our requirements (or NAAS is able to fix the files to meet our requirements), the figure will be marked as "fixed" above. If NAAS is unable to fix the files, a red "failed" label will appear above. When NAAS has confirmed that the figure files meet our requirements, please download the file via the download option, and include these NAAS processed figure files when submitting your revised manuscript.-->

Statistical Learning for Climate-GDP Panels: Data Cleaning, Flexible Trend Controls, and Predictive Validation

PCLM-D-26-00056R1

Dear Dr. Schötz,

We are pleased to inform you that your manuscript 'Statistical Learning for Climate-GDP Panels: Data Cleaning, Flexible Trend Controls, and Predictive Validation' has been provisionally accepted for publication in PLOS Climate.

Before your manuscript can be formally accepted you will need to complete some formatting changes, which you will receive in a follow-up email from a member of our team.

Please note that your manuscript will not be scheduled for publication until you have made the required changes, so a swift response is appreciated.

IMPORTANT: The editorial review process is now complete. PLOS will only permit corrections to spelling, formatting or significant scientific errors from this point onwards. Requests for major changes, or any which affect the scientific understanding of your work, will cause delays to the publication date of your manuscript.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they'll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact climate@plos.org.

Thank you again for supporting Open Access publishing; we are looking forward to publishing your work in PLOS Climate.

Best regards,

María Dolores Gadea Rivas

Academic Editor

PLOS Climate

***********************************************************

Additional Editor Comments (if provided):

The authors have satisfactorily addressed the referee’s comments, and the paper is recommended for acceptance

Reviewer Comments (if any, and for reference):

Reviewer's Responses to Questions

-->Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.-->

Reviewer #1: All comments have been addressed

**********

-->2. Does this manuscript meet PLOS Climate’s publication criteria? Is the manuscript technically sound, and do the data support the conclusions? The manuscript must describe methodologically and ethically rigorous research with conclusions that are appropriately drawn based on the data presented.-->

Reviewer #1: Yes

**********

-->3. Has the statistical analysis been performed appropriately and rigorously?-->

Reviewer #1: Yes

**********

-->4. Have the authors made all data underlying the findings in their manuscript fully available (please refer to the Data Availability Statement at the start of the manuscript PDF file)?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception. The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.-->

Reviewer #1: Yes

**********

-->5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS Climate does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.-->

Reviewer #1: Yes

**********

-->6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)-->

Reviewer #1: (No Response)

**********

-->7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

Do you want your identity to be public for this peer review?  If you choose “no”, your identity will remain anonymous but your review may still be made public.

For information about this choice, including consent withdrawal, please see our Privacy Policy.-->

Reviewer #1: Yes: Alvaro Escribano

**********