Forecasting Regional Tourism Demand in Morocco from Traditional and AI-Based Methods to Ensemble Modeling
Round 1
Reviewer 1 Report
My main issue regarding the paper is that there is no explanation how the forecasts were obtained, which models were used . For example, it is not enough to say an Ar model. Which AR? Therefore I feel that there are reproducibility issues here.
Also why invert the colors of the trining and test sets bewteen Figures 5 and 6? It makes life difficult for the reader.
Author Response
We appreciate the time and effort that you have dedicated to providing your valuable feedback on our manuscript.
We have been able to incorporate changes to reflect all the comments and the suggestions you provided. We have used the “Track Changes” function in word to highlight the incorporated changes within the manuscript.
Here is a point-by-point response to the reviewer’s comments. See the attachment and the Manuscript.
We would like to thank you again for taking the time to review our manuscript.
Best Regards.
Author Response File: 
Author Response.docx
Reviewer 2 Report
The study examines the forecasting performance of the AI-based models against the traditional time series models. Methodologically, there is no significant contribution to the literature as all models have been examined in the published studies. In addition, when yu are carrying out traditional time series analysis, you need to examine the properties of the time series (order of integration) to be able to correctly specify the models for forecasting, which is absent from the study. In addition, the literature currently focuses on searching for the appropriate forecasting models/strategy in the time of crisis, such as the COVID-19. An effort should be made in this aspect, i.e., to extend the data to include the data during the COVID-pandemic. This will make your study more relevant and generate bigger impact both for research and practice.
Author Response
We appreciate the time and effort that you and the reviewers have dedicated to providing your valuable feedback on our manuscript.
We have been able to incorporate changes to reflect most of the suggestions provided by the reviewers. We have used the “Track Changes” function in word to highlight the incorporated changes within the manuscript.
Here is a point-by-point response to the reviewer’s comments. See the Attachment.
We would like to thank you again for taking the time to review our manuscript.
Best Regards.
Author Response File: Author Response.docx
Reviewer 3 Report
Please refer to the attached report.
Comments for author File: 
Comments.pdf
Author Response
We appreciate the time and effort that you and the reviewers have dedicated to providing your valuable feedback on our manuscript.
We have been able to incorporate changes to reflect most of the suggestions provided by the reviewers. We have used the “Track Changes” function in word to highlight the incorporated changes within the manuscript. See the attachment.
We would like to thank you again for taking the time to review our manuscript.
Best Regards.
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
Thank you for the revised manuscript. The authors have addressed some of my comments, but they cannot extend the data to include the the observations over the pandemic period. I understand the difficulty in obtaining the data, you may therefore include this as a limitation and provide suggestions for future research when the data become avaliable. In fact some of the recent published studies on tourism demand forecasting did use the data for 2021.
Author Response
Please see the attachment
Author Response File: Author Response.docx
Reviewer 3 Report
The Authors have carefully addressed all comments contained in the first review report and have provided the required details.
In this version, my main concerns relate both to the dataset and to the appropriateness of forecasting techniques. The Authors only use 19 annual observations, which is a very small sample size for the implementation of either traditional or AI-based forecasting methods. Furthermore, the dataset must be split into training and test set, with even smaller sample sizes.
If I understand correctly, the data points correspond to annual numbers of tourist arrivals (aggregating all seasons of the year): in my opinion, this approach cannot generate timely and useful forecasts for policy decisions and revenue management actions. I would personally require a higher-frequency dataset with a much larger sample size.
Author Response
Please see the attachment
Author Response File: Author Response.docx
