Do earnings matter for stock prices? Evidence from the BVL using Event Study methodology and Data Science
DOI:
https://doi.org/10.71701/zen3y225Keywords:
Event Study, Data Science, BVL, EPU ETF, ADR, Abnormal returnsAbstract
This study examines the impact of quarterly earnings announcements of major Peruvian American Depositary Receipts (ADRs) on stock returns, applying the Event Study methodology within a Data Science framework. The research is applied, quantitative, explanatory, and follows a non-experimental longitudinal design covering the period 2015–2024. The dataset comprises 2,514 daily adjusted price observations for each stock obtained from Yahoo Finance, along with financial results sourced from Alpha Vantage through R programming and API integration.
The stylized facts show that returns exhibit a symmetric distribution centered around zero, display fat tails, and generally follow autoregressive processes of order 2. Based on MacKinlay’s (1997) framework, the event analyzed corresponds to quarterly earnings announcements. Expected returns were estimated using the market model, with the EPU ETF serving as a proxy for the Peruvian market. The estimation window consisted of 250 trading days prior to the event, while the event window and post-event window were set at one day and ten days, respectively.
A Data Science approach was incorporated to refine the threshold for classifying earnings surprises (actual vs. expected), setting it at 5% rather than the traditional 2.5%, given the higher sensitivity of emerging markets to political and social shocks. A total of 144 events were modeled through regression analysis. Three stocks showed statistically significant results across all specifications, while the fourth did so in 67% of the cases. BAP was identified as the least risky asset and BVN as the riskiest.
The findings indicate that positive earnings announcements generate an average abnormal return of 0.87%, while the effects of negative and neutral announcements remain statistically inconclusive. On average, positive events reach their price peak within three days, while negative events reach their minimum within five days.
Downloads
References
[1] Campbell, J., Lo, A., & MacKinlay, A. (1997). The econometrics of financial markets. Princeton University Press.
[2] Chapman, P. C. (2000). CRISP-DM 1.0: Step-by-step data mining guide. CRISP-DM Consortium.
[3] Chen, J. (2025, 16 septiembre). Exchange-Traded Fund (ETF): What It Is and How to Invest. Investopedia. https://www.investopedia.com/terms/e/etf.asp
[4] Corrado, C. (2010). Event studies: A methodology review. Accounting and Finance, 50(1), 207–234.
[5] Fuenzalida, D., Mongrut, S., & Nash, M. (2008). Stock split en la Bolsa de Valores de Lima: ¿Afectan el rendimiento y la liquidez de los títulos? Estudios Gerenciales, 24(107), 11–36.
[6] Galindo, H., & Montesinos, A. (2018). Macroeconomía dinámica. Fondo Editorial EDUNI.
[7] García Gutiérrez, S. (2012). El contenido informativo de los anuncios de dividendos y la reacción del precio de las acciones: Perú 2001-2010 [Tesis de maestría, Universidad Nacional Mayor de San Marcos]. Cybertesis UNMSM.
[8] Hernandez, M. (2002). Macroeconomic reform and policy: The case of Peru – Analyzing the effects of some major economic, political and social changes in the Peruvian financial market for the period between 1990 and 1992. Undergraduate Journal of Economics, 7(1), Art. 4.
[9] Huntington-Klein, N. (2022). The effect. CRC Press.
[10] Hyndman, R., & Athanasopoulos, G. (2018). Forecasting: Principles and practice. OTexts.
[11] IBM. (2021, marzo 4). IBM SPSS Modeler. https://www.ibm.com/docs/en/spss-modeler/18.0.0?topic=spss-modeler-crisp-dm-guide
[12] MacKinlay, A. (1997). Event studies in economics and finance. Journal of Economic Literature, 35(1), 13–39.
[13] Melgarejo, M., Montiel, E., & Sanz, L. (2016). The stock market’s reaction to accounting information: The cases of Chile and Peru. Journal of Accounting in Emerging Economies, 6(3), 254–268.
[14] Porras Cerrón, J. (2017). Pruebas no paramétricas usando R. Universidad Nacional Agraria La Molina.
[15] Rocca Carbajal, L. (2017). El mercado de valores en fácil. Pontificia Universidad Católica del Perú.
[16] Salas, A. (2021). Cambios no anticipados positivos y negativos de la tasa de interés de referencia y la rentabilidad del índice general de la Bolsa de Valores de Lima entre los años 2003 y 2019 [Tesis de título profesional, Universidad Nacional de Ingeniería].
[17] Soto, I., & Gamboa, J. (2021). Ciencia de datos con R: Métodos estadísticos para la investigación experimental. Universidad Nacional Agraria La Molina.
[18] Tocón Vega, D. (2016). ¿El cambio de CEO puede afectar el valor de las firmas?: Un análisis de estudio de eventos en el mercado bursátil peruano [Tesis de grado, Universidad Peruana de Ciencias Aplicadas].
[19] Tsay, R. (2010). Analysis of financial time series. Wiley.
[20] Zivot, E. (2021). Introduction to computational finance and financial econometrics with R. https://bookdown.org/
[21] Zumel, N., & Mount, J. (2020). Practical data science with R. Manning Publications.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Roberto León Leyva (Autor/a)
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
