Addition of Subset and Dummy Variables in the Threshold Spatial Vector Autoregressive with Exogenous Variables Model to Forecast Inflation and Money Outflow

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The paper presents a valuable enhancement to the TSpVARX model by incorporating subset and dummy variables to improve forecasting accuracy for inflation and money outflow. The methodological rigor and relevance of the topic make it a noteworthy contribution to econometrics and economic forecasting. However, there are a few areas that need more clarification and improvements for broader accessibility and understanding.

1 Introduction:

• • Clarification of Model Choice: Provide a more explicit justification for choosing TSpVARX over other advanced models. This can include a brief comparison of TSpVARX’s unique advantages in handling non-linear relationships.
  • Explanation of Dummy and Subset Variables: Simplify the discussion on the addition of dummy and subset variables to help readers understand their importance in the model.
  • Highlight Practical Applications: The introduction could benefit from briefly mentioning potential real-world applications, such as policy impact assessments, inflation control, or economic planning. This will help to frame the study's relevance from the outset.

2 Materials and Methods:

• • Provide a High-Level Summary: Before diving into the detailed methodology, offer a brief conceptual overview explaining each step's purpose. For example, "This section explains the steps taken to enhance the TSpVARX model to better capture economic fluctuations influenced by events and seasonality."
  • Explain Variable Roles in Simple Terms: Clarify why dummy and subset variables are critical for the model's forecasting capability. For instance, "The dummy variables enable the model to account for events like fuel price changes, while the subset variables capture recurring seasonal effects."
  • Dataset Justification: Include a short explanation of why Yogyakarta, Solo, and Semarang were chosen. Mention if these regions have specific economic characteristics relevant to the model, which will add depth to the methodological choice.

3 Results:

• • Add Visual Comparisons: Including charts or graphs that illustrate how TSpVARX performs with and without the subset and dummy variables could enhance the reader’s understanding and highlight the model's effectiveness.
  • Simplify Statistical Terms: Provide brief explanations of key statistical metrics, such as RMSE, in a way that a broader audience can understand. For example, "RMSE measures the model’s prediction accuracy. A lower RMSE indicates more precise predictions."
  • Interpret Results in Practical Terms: Rather than focusing solely on statistical metrics, briefly explain what the improvements in accuracy mean for real-world applications, like more reliable inflation predictions that can aid policymakers.

4 Discussion:

• • Highlight Model Limitations and Future Research: While the discussion covers the model’s advantages, it would be beneficial to explicitly mention any limitations or constraints in the model's application, especially regarding unique economic contexts.
  • Outline Practical Implications: Describe how the model’s enhanced forecasting can aid policymakers or economists. For example, "Improved accuracy in inflation forecasting helps central banks adjust monetary policy with greater confidence."
  • Suggest Specific Extensions: Recommend areas for future research, such as testing the model with different types of economic events or in diverse regional contexts, to broaden its applicability.

5 Conclusion:

• • Summarize the Model's Main Contributions: Clearly state how the addition of subset and dummy variables advances inflation and money outflow forecasting, providing a final summary of the study's contributions.
  • Reinforce Practical Value: Reiterate the real-world relevance of the findings, emphasizing how policymakers and analysts can use the improved TSpVARX model for effective economic planning.
  • Provide Direction for Future Research: Suggest specific next steps, such as applying the model to different economic indicators or regions, which can help to inspire further research on the topic.

Comments on the Quality of English Language

While the English quality is understandable, an expert English editor could help refine the text, ensuring clarity and consistency. This would especially benefit sections with complex technical explanations and improve the overall flow, making the paper more accessible to an international readership.

Example 1

Current Text: "The TSpVARX model can accommodate simultaneously the mutual relationship between variables, the influence of metric exogenous variables, the influence of lag variables of endogenous variables, inter-regional linkages, and the nonlinearity between endogenous variables and predetermined variables."

Suggested Improvement: "The TSpVARX model captures complex interactions by accounting for relationships between variables, external factors, past data points, regional connections, and nonlinear patterns."

Example 2

Current Text: "The subset variable used is the 12th lag of the endogenous variable, while the dummy variable used is the variable of the increase and decrease in fuel prices."

Suggested Improvement: "We use a 12th lag subset variable to capture seasonal effects and a dummy variable to represent fuel price changes."

Example 3

Current Text: "This study proposes the addition of subset and dummy variables to improve the TSpVARX model’s forecasting ability, particularly for predicting inflation and money outflow, by accommodating both recurring seasonal patterns and specific events such as fuel price changes."

Suggested Improvement: "This study improves the TSpVARX model by adding subset and dummy variables. These additions enhance the model's accuracy in forecasting inflation and money outflow by accounting for recurring patterns and specific events like fuel price changes."

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The study is up-to-date and interesting because it addresses complex economic relationships using modern modeling techniques (TSpVARX) to forecast inflation and monetary flows in real economic contexts, such as seasonal events and fuel price adjustments, providing valuable insights for monetary policy and emerging economies.

My specific observations are:

1. Although the abstract is clear and provides a succinct description of the background, method and conclusions, it does not state the main purpose of the study. Is it just to improve forecast accuracy or also to gain a deeper understanding of the relationships between variables?
2. How is the fluctuating forecast better than the original one? Is the desired fluctuation a primary objective or just a side effect?
3. It might be useful to clarify how each pre-existing model is an intermediate step towards TSpVARX.
4. Why is lag 12 chosen to represent the Eid al-Fitr event?
5. How are dummy variables concretely integrated into the proposed model? It is not clear how these variables impact the relationship between endogenous and predetermined variables?
6. How are the dummy variables concretely integrated into the proposed model? Is it not clear how these variables impact the relationship between endogenous and default variables?
7. Some phrases used in the methodology section are very technical and dense. Additional definitions or rewording could be useful to make the text more accessible to non-specialized readers. For example, explanation of subset and dummy variables.
8. Line 82 states that the forecast will be carried out "from June 2003 to December 2024" ...should be June 2023.
9. In step 6 of subsection 2.5, the use of the RESET test is mentioned. A brief explanation of this test would be helpful.
10. What are the rationales for choosing Yogyakarta, Solo and Semarang regions? Are there regional particularities that justify their selection?
11. As there are many formulas and technical details, the inclusion of flow charts could improve readability and understanding.
12. How are forecast results validated?
13. Not all figures are accompanied by concise and relevant explanations. For example, Figures 7 and 8.
14. In the Conclusions section we reiterate the most important observations, such as significantly lower RMSE values or improved forecasts for certain events.
15. How the study contributes to the existing literature and how it can be used in practice.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Thanks for the responses to the comments I made. I believe that in this new form the article can be published.