Transformaciones logarítmicas en regresión simple
Logarithmic transformations in simple regression analysis
Abstract (en)
In this paper the effect of the logarithmic transformations in simple regression analysis is investigated. In practice, it is very common that exponential and power models’ parameters are estimated by means of a logarithmic transformation which reduces them to a linear form. The estimations in the initial models are obtained by applying the exponential function to the intercept estimation. In this work, it is found that this procedure does not generate least squares solutions for the initial model and introduces variations in the way in which relationships between variables are conceived. Because of the popularity of software tools, the risk of using inappropriate models for the data may be unnoticed.
Abstract (es)
En este artículo se investiga los efectos de las transformaciones logarítmicas en un análisis de regresión simple. En la práctica, es muy común que los parámetros de los modelos conocidos como exponencial y potencial se estimen de manera habitual mediante una transformación logarítmica, que los reduce a modelos lineales y se “regresa” al modelo original aplicando la función exponencial a la estimación del intercepto. En este trabajo se encuentra que este procedimiento no genera estimadores de mínimos cuadrados para el modelo inicial e introduce variaciones en la forma como se conciben las relaciones entre las variables. La popularidad de las herramientas de análisis hace que el riesgo de utilizar modelos que no correspondan a los datos pase desapercibido.
References
Mendenhall, W. & McClave, J. (1981), A Second Course in Business Statistics: Regression Analysis, Dellen Publishing Company, Santa Clara, California.
Walpole, R., Myers, R., Myers, S. & Ye, K. (2012), Probability & Statistics for Engineers & Scientists, Prentice Hall, New York.
How to Cite
License
The authors maintain the rights to the articles and therefore they are free to share, copy, distribute, execute and publicly communicate the work under the following conditions:
Recognize the credits of the work in the manner specified by the author or licensor (but not in a way that suggests that, you have their support or that they support your use of their work).
Comunicaciones en Estadística is licensed under Creative Commons Atribución-NoComercial-CompartirIgual 4.0 Internacional (CC BY-NC-SA 4.0)
Universidad Santo Tomás preserves the patrimonial rights (copyright) of the published works, and favors and allows the reuse of them under the aforementioned license.