Publication detail

Boscovich fuzzy regression line

ŠKRABÁNEK, P. MAREK, J. POZDÍLKOVÁ, A.

English title

Boscovich fuzzy regression line

Type

journal article in Web of Science

Language

en

Original abstract

We introduce a new fuzzy linear regression method. The method is capable of approximating fuzzy relationships between an independent and a dependent variable. The independent and dependent variables are expected to be a real value and triangular fuzzy numbers, respec-tively. We demonstrate on twenty datasets that the method is reliable, and it is less sensitive to outliers, compare with possibilistic-based fuzzy regression methods. Unlike other commonly used fuzzy regression methods, the presented method is simple for implementation and it has linear time-complexity. The method guarantees non-negativity of model parameter spreads.

English abstract

We introduce a new fuzzy linear regression method. The method is capable of approximating fuzzy relationships between an independent and a dependent variable. The independent and dependent variables are expected to be a real value and triangular fuzzy numbers, respec-tively. We demonstrate on twenty datasets that the method is reliable, and it is less sensitive to outliers, compare with possibilistic-based fuzzy regression methods. Unlike other commonly used fuzzy regression methods, the presented method is simple for implementation and it has linear time-complexity. The method guarantees non-negativity of model parameter spreads.

Keywords in English

fuzzy linear regression; non-symmetric triangular fuzzy number; least absolute value; Boscovich regression line; outlier

Released

23.03.2021

Publisher

MDPI

Location

Basel, Switzerland

ISSN

2227-7390

Volume

9

Number

6

Pages from–to

1–14

Pages count

14

BIBTEX


@article{BUT171143,
  author="Pavel {Škrabánek} and Jaroslav {Marek} and Alena {Pozdílková},
  title="Boscovich fuzzy regression line",
  year="2021",
  volume="9",
  number="6",
  month="March",
  pages="1--14",
  publisher="MDPI",
  address="Basel, Switzerland",
  issn="2227-7390"
}