Comparison of Robustness Non-Linearity Test in Computational Statistics when Outlier Detected.

Saved in:

Bibliographic Details
Title:	Comparison of Robustness Non-Linearity Test in Computational Statistics when Outlier Detected.
Authors:	Rantini, Dwi¹ dwi.rantini@ftmm.unair.ac.id, Ramadan, Arip² aripramadan@telkomuniversity.ac.id, Sesay, Alhassan³ alhassansesay3@gmail.com
Source:	Engineering Letters. Dec2024, Vol. 32 Issue 12, p2299-2323. 25p.
Subjects:	Computational statistics, Regression analysis, Statistics
Abstract:	In regression modeling, we often encounter data with different ranges. Such data usually have outliers. If an outlier has a value far from the mean, it can cause an error in modeling. For example, data has a quadratic pattern, but because there are outliers, it can be indicated that the data is linear. This research will prove which non-linearity test is more robust if outlier data is shown. To prove this, data is generated, and outliers are found in the variable response of the nonlinear model. Using the RESET, the Terasvirta and White tests will prove to be more robust. The results show that the Terasvirta test is more robust than the RESET and White tests. This statement applies to models that are non-linear in parameters and variables. Therefore, if we want to test the goodness of a non-linear model and outliers are detected from our research, we recommend using the Terasvirta test. We prove that 53.42% of Terasvirta performs better than the RESET and White tests. Because the Terasvirta test is proven to be more robust if outliers are found in the non-linear model, this is very important to increase knowledge in education, especially in computing statistics. [ABSTRACT FROM AUTHOR]
	Copyright of Engineering Letters is the property of International Association of Engineers (IAENG) and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database:	Engineering Source

Description
Abstract:	In regression modeling, we often encounter data with different ranges. Such data usually have outliers. If an outlier has a value far from the mean, it can cause an error in modeling. For example, data has a quadratic pattern, but because there are outliers, it can be indicated that the data is linear. This research will prove which non-linearity test is more robust if outlier data is shown. To prove this, data is generated, and outliers are found in the variable response of the nonlinear model. Using the RESET, the Terasvirta and White tests will prove to be more robust. The results show that the Terasvirta test is more robust than the RESET and White tests. This statement applies to models that are non-linear in parameters and variables. Therefore, if we want to test the goodness of a non-linear model and outliers are detected from our research, we recommend using the Terasvirta test. We prove that 53.42% of Terasvirta performs better than the RESET and White tests. Because the Terasvirta test is proven to be more robust if outliers are found in the non-linear model, this is very important to increase knowledge in education, especially in computing statistics. [ABSTRACT FROM AUTHOR]
ISSN:	1816093X