Methodology for Estimating the Cost of Capital (Ke)

Methodological Update: From Excel to Python in Estimating the Cost of Capital

Critical Review as a Driver of Improvement

At Novagestión Asesorías Financieras SpA, professional excellence is built on the constant practice of critically reviewing our own procedures. We do not assume that what is established is optimal; on the contrary, we subject our methodologies to rigorous scrutiny, questioning assumptions and exploring new tools that enhance the quality of our analyses and recommendations.

he methodological revision we present stems from an internal conviction: the duty to update our procedures to keep them aligned with the state of the art in quantitative finance, balancing accumulated knowledge with the capabilities offered by new technologies.

From Excel to Python: A Quantitative Leap

Our previous methodology, based on Excel spreadsheets, served its purpose for years, allowing us to estimate the cost of capital (Ke) with a practical and transparent approach. However, the growth in data volume, the need for longer time series, and the demand for reports with greater statistical depth revealed its limitations.

The challenge was clear: to preserve the conceptual soundness of Damodaran's model while moving it to a more robust, scalable, and reproducible environment. Choosing Python allowed us to implement procedures that were unfeasible in Excel: numerical optimization (SLSQP), Monte Carlo simulation (jump-diffusion), root search (Brentq), and automatic generation of advanced statistical graphs.

Balance Between Tradition and Innovation

The update did not mean abandoning accumulated knowledge, but rather enhancing it. We preserved the theoretical foundations that guide our practice:

• ✅  The Damodaran equation as the core of the model.
• ✅  Market risk-free rates (BCP/BCU) as long-term anchors.
• ✅  The distinction between nominal and real results.
• ✅  The end-of-month filter to reduce noise.

And we integrated new knowledge:

• ✅  Numerical optimization with convex constraints.
• ✅  Stochastic modeling of the growth rate using jump-diffusion.
• ✅  Evaluation of distributions using the Kolmogorov-Smirnov test and QQ plots.
• ✅  Full automation to ensure reproducibility and reduce human error.

Challenges Embracing to Deliver Quality

Each update brings challenges that we embrace responsibly:

• ✅   Technical: Implementing intensive processes without sacrificing transparency. Every step is documented so that the team and clients understand the "how" and the "why."
• ✅   Conceptual: Maintaining consistency between ex-ante and ex-post, nominal and real approaches, presenting results across multiple dimensions for a comprehensive view.
• ✅   Validation: Subjecting new results to consistency tests with the previous methodology, ensuring that any differences reflect technical improvements.
• ✅   Service: Translating technical complexity into useful information for investment and valuation decisions.

Commitment to Continuous Improvement

This publication is a milestone on our continuous improvement journey, not an end point. Financial knowledge and technological tools are constantly evolving, and we are committed to staying at the forefront of both.

We invite colleagues, clients, and peers to review this methodology, question it, and enrich it. Constructive criticism is the fuel of progress, and at Novagestión we value it as an asset.

The service we offer is measured not only by the quality of our responses, but also by the rigor of our questions. This update is the result of a question we asked ourselves: Can we do better? The answer is a resounding yes.