Statistical Modeling: A New Regression Curve Approximation using Mixed Estimators Truncated Spline and Epanechnikov Kernel

Sifriyani, Andrea Tri Rian Dani, Meirinda Fauziyah, I. Nyoman Budiantara

Research output: Contribution to journalArticlepeer-review

Abstract

In the era of computation, researchers have paid significant attention to the nonparametric regression method. Nonparametric regression has the benefit of a high degree of modeling flexibility. Developing a mixed estimator truncated spline-Epanechnikov kernel is the most recent innovation in this study. The truncated spline estimator excels at handling data whose behavior varies at predetermined intervals. In contrast, the Epanechnikov kernel estimator has a more flexible structure and excels at modeling data that does not adhere to a particular pattern. Maximum Likelihood Estimation is utilized to estimate parameters. The concluding section of this study will discuss the estimator properties of the mixed estimators truncated spline and Epanechnikov kernel models. The proposed can be utilized for additional analysis in the field of nonparametric regression.

Original languageEnglish
Pages (from-to)1649-1655
Number of pages7
JournalEngineering Letters
Volume31
Issue number4
Publication statusPublished - 1 Nov 2023

Keywords

  • Epanechnikov Kernel
  • Maximum Likelihood Estimation
  • Mixed Estimators
  • Nonparametric Regression
  • Truncated Spline

Fingerprint

Dive into the research topics of 'Statistical Modeling: A New Regression Curve Approximation using Mixed Estimators Truncated Spline and Epanechnikov Kernel'. Together they form a unique fingerprint.

Cite this