Article

A new method for detecting outliers in Data Envelopment Analysis

---

Data Envelopment Analysis (DEA) is a computationally convenient way to measure efficiency that does not require an explicit functional relationship between inputs and outputs. However, because the frontier is constructed using extreme observations, DEA can be sensitive to extreme points, especially when data may be contaminated by measurement error. In such settings, the technical efficiency scores calculated from datasets that include outliers can be misleading. Timmer (1971) was among the first to recognize the potential sensitivity of computed technical efficiency measures to outliers when linear programming techniques are used to measure efficiency. In other settings, outliers may simply represent outcomes observed accurately, but with low frequency, and hence, worthy of further investigation

---

Ketersediaan

Informasi Detail

Judul Seri

A new method for detecting outliers in Data Envelopment Analysis

No. Panggil

ATC LO TRA a

Penerbit

United States of America : Routledge., 2008

Deskripsi Fisik

4p

Bahasa

English

ISBN/ISSN

1466 4291

Klasifikasi

LO

Tipe Isi

-

Tipe Media

-

Tipe Pembawa

online resource

Edisi

2008

Subjek

Data envelopment analysis
detecting outliers
A new method for detecting outliers

Info Detail Spesifik

-

Pernyataan Tanggungjawab

Nam Anh Tran

Versi lain/terkait

Lampiran Berkas

Komentar

---

Anda harus masuk sebelum memberikan komentar