Published
2008-01-01
UNA NUEVA PRUEBA PARA EL PARÁMETRO DE DIFERENCIACIÓN FRACCIONAL
A NEW TEST FOR THE FRACTIONAL DIFFERENCING PARAMETER
Keywords:
memoria larga, modelo ARFIMA, aproximación autorregresiva, identificación, prueba de hipótesis, diferencia fraccional (es)Long memory, Arfima model, Autoregressive process, Identification, Testing hypothesis, Fractional differencing (en)
Downloads
Este documento presenta una nueva prueba para el parámetro de diferenciación fraccional de un modelo ARFIMA, basada en una aproximación autorregresiva de su componente a corto plazo. El comportamiento de la prueba se estudia por medio de experimentos Monte Carlo en una distribución normal, y se compara con el comportamiento de algunas de las pruebas más utilizadas. Para los casos estudiados, se concluye que la nueva prueba tiene generalmente potencias superiores, conservando un tamaño adecuado. A partir de la estimación del parámetro de diferenciación fraccional usando el modelo aproximado, es posible identificar el modelo correcto para la componente a corto plazo, lo cual permite mejorar la inferencia sobre dicho parámetro. Una ventaja adicional del procedimiento propuesto es que permite probar la existencia de larga memoria en presencia de errores dependientes, como en el caso de modelos de volatilidad de la familia ARCH. Se ilustra su aplicación en un procedimiento de identificación y estimación de un modelo ARFIMA–ARCH usando datos simulados.
This paper presents a new test for the fractional differencing parameter of an ARFIMA model, based on an autoregressive approximation of its short-range component. The test’s behavior is studied using Monte Carlo simulations under a normal distribution and is compared to results found for others well–known long memory tests. In general, the results show that the new test has a superior power while maintaining an adequate size of the test. From the estimation of the fractional differencing parameter using the approximate model, it is possible to identify the correct model for the short–term component, which allows improving the inference on the above mentioned parameter. An additional advantage of the proposed procedure is the possibility of testing long memory in the presence of dependent errors such as in the volatility models of ARCH family. The identification and estimation procedure is applied to simulated data from an ARFIMA–ARCH model.
Article abstract page views
624
Downloads
Download data is not yet available.
How to Cite
Castaño, E., Gómez, K., & Gallón, S. (2008). A NEW TEST FOR THE FRACTIONAL DIFFERENCING PARAMETER. Revista Colombiana De Estadística, 31(1), 67-84. https://revistas.unal.edu.co/index.php/estad/article/view/29594
License
Copyright (c) 2008 Revista Colombiana de Estadística

This work is licensed under a Creative Commons Attribution 4.0 International License.
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).






