As partes interessadas estão questionando seus modelos econômicos. Como você defenderá sua validade?

To defend the validity of our economic models, I would start by explaining the foundational frameworks and well-established theories they are based on. Highlight the reliability and robustness of the data sources used. Next, I would describe the assumptions made and validate them with credible evidence and industry norms. Additionally, demonstrating past successes and outcomes where the models accurately predicted results will reinforce their credibility. Finally, I would offer to run sensitivity analyses to show how the models perform under different scenarios, underscoring their adaptability and reliability.

In situations like this, I focus on three key strategies: 1. Transparent Methodology: I clearly explain the data sources, assumptions, and methodologies used in the models. By walking stakeholders through the step-by-step process, they can see the rigorous foundation on which the models are built. 2. Evidence of Accuracy: I will present historical data and case studies where similar models have accurately predicted outcomes. 3. Open to Feedback: I actively listen to stakeholders' concerns and am open to re-evaluating certain aspects of the model if new, relevant data is presented.

1. You can bring various case studies to your defence. All that literature survey for your master's/doctorate thesis will come into play here. 2. You can bring forth similar cases that you've worked on in the past and map the trajectory of that process in your defence. If you were able to produce good results before, chances are you'll be able to do that once again.

To defend the validity of our economic models, I will demonstrate the robustness of the data sources and the underlying assumptions used. I'll highlight the rigorous methodologies and statistical techniques employed, such as sensitivity analysis and back-testing, to ensure the models' accuracy and reliability. Additionally, I'll present historical performance comparisons and case studies where the models accurately predicted outcomes. Engaging in transparent discussions about potential limitations and continuously updating the models based on new data and feedback will further reinforce their credibility.

It is recognised by academics and partitioners that when the model assumptions do not work in practice, they need to be relaxed to improve the accuracy and applicability of the models. Sometimes, assumptions are not critical and relaxing them can lead to more accurate models. E.g., the assumption of rationality in traditional economic models has been challenged by behavioural economics. This has led to the development of more realistic models that better capture human behaviour. Another example is the assumption of perfect information. In reality, information is often incomplete or imperfect, leading to market inefficiencies. This has led to the development of models that incorporate information asymmetry, such as the principal-agent model.

In defending the validity of economic models amid stakeholder skepticism, it's crucial to emphasize that assumptions are essential simplifications, not flaws. They allow us to isolate and study specific economic phenomena. These assumptions are meticulously derived from empirical data and past behaviors, ensuring relevance. Acknowledge that models are not static; they evolve as new information emerges. By remaining transparent and flexible, adapting assumptions to reflect real-world changes, we reinforce the model's robustness, making it a reliable tool for decision-making.

It is common in the professional life that various stakeholders (management, governmental officials, the expert community, and journalists) would question the validity of economic models. So we have to deal with it. Many economic modellers are natural introverts. It is hard to talk to people, but we have to do it! It helps in practice to allocate at least 25% of working time for an outreach. Since we want our models to have a real-life impact, we have to go out and present them, discuss their logic and assumptions, get feedback, and present again. These questions would help improve modeling. So, my best advice would be to consciously allocate 25% of working time (at the very least, preferably more!) for discussions and presentations.

1. Assumptions: Clearly outline the assumptions behind the model. Explain why these assumptions were made and how they align with industry standards or empirical evidence. 2. Data Sources: Provide detailed information about the data used in the model. Highlight the credibility of the data sources and any steps taken to ensure accuracy and relevance. 3. Methodology: Walk stakeholders through the methodology, explaining the rationale for the chosen approach. Emphasize any best practices or standard techniques used in economic modeling.

Defending the methodological rigor of your models is something every economist has to deal with occasionally. When I was writing my dissertation on evaluating the impact of quarterly EPS on CEO dismissal using econometrics and natural experiments, I explained how the model was grounded in established economic theory, ensuring it aligns with the fundamental principles governing the variables in question. This approach serves as the foundation for your projects. Next, I focused on highlighting the robustness of the data used, including steps taken for data pre-processing and transformation. Finally, I verified my model by looking at industry standards. Using this three-step approach convinced my committee that my project was well founded.

Demonstrate model robustness through rigor 1. Sensitivity Analysis: Show how the model behaves under different scenarios by conducting a sensitivity analysis. This helps illustrate that the model’s conclusions are reliable even when key variables are adjusted. 2. Validation Against Historical Data: If possible, demonstrate that the model has been validated against historical data or other reliable benchmarks. Show how well the model’s predictions align with actual outcomes. 3. Comparisons with Other Models: Compare your model’s results with those from other established models or approaches. This can help show that your model is consistent with broader economic thinking.

-Data quality: Ensure data accuracy, completeness, and reliability. -Data sources: Clearly document the sources of all data used in the model. -Data validation: Explain the methods used to validate data quality.

Thorough documentation is an integral part of building a robust economic model. Documentation should include data sources, methodology, and, in some cases, a user manual. A well-documented model creates trust and facilitates easier use.

-Backtesting: Demonstrate the model's performance in predicting past events. -Case studies: Provide examples of how the model has been successfully applied in real-world scenarios. -Practical relevance: Highlight the model's ability to inform decision-making and generate actionable insights.

1. Demonstrate Practical Use Cases: Show how the model has been or can be used to inform decisions, improve outcomes, or solve problems. Practical examples can help stakeholders see the model’s value. 2. Show Impact on Decision-Making: Explain how the model’s results will influence decisions and lead to positive outcomes for the organization or community.

Be open to revisions through iterative improvement: Show that you are open to revising and improving the model based on new data, feedback, or changing circumstances. This demonstrates flexibility and a commitment to accuracy.

-Iterative process: Emphasize that model development is an ongoing process involving refinement and updates. -Feedback loop: Encourage stakeholder feedback and incorporate it into model improvements. -Transparency: Communicate changes to the model and their impact on results.