Full-text resources of CEJSH and other databases are now available in the new Library of Science.
Visit https://bibliotekanauki.pl

Results found: 5

first rewind previous Page / 1 next fast forward last

Search results

Search:
in the keywords:  Robust statistical methods
help Sort By:

help Limit search:
first rewind previous Page / 1 next fast forward last
1
Publication available in full text mode
Content available

Wielokryterialne porządkowanie metodą PROMETHEE odporne na zmiany wag kryteriów

100%
Sobotka O.
Studia Ekonomiczne
|
2013
|
vol. 163
209-219
EN
The approach described in article is based on the Linear Partial Information (LPI) by E. Kofler. The robust multicriteria ranking is based on the robust dominance relations, which holds under all admissible combinations of criteria weights. In the article the robust dominance relation is defined and the robust ranking method based on the PROMETHEE outranking relations is proposed.
2
Publication available in full text mode
Content available

Robust Ordinal Regression Applied to TOPSIS

100%
Zielniewicz P.
Multiple Criteria Decision Making
|
2013
|
vol. 8
178-196
EN
This paper proposes a new method for ranking a nite set of alternatives evaluated on multiple criteria. The presented method combines the robust ordinal regression (ROR) approach and the ranking score based on the aggregate distance measure function coming from the TOPSIS method. In our method, the preference model is a set of ad- ditive value functions compatible with a non-complete set of pairwise comparisons of some reference alternatives given by the decision maker (DM). Based on this set of compatible value functions, we dene an ag- gregate function representing relative closeness to the reference point (ideal solution) in the value space. The ranking score determined by this distance measure is then used to rank all alternatives. Calculating the distance in the value space permits to avoid normalization used in TOPSIS to transform original evaluations on different criteria scales into a common scale. This normalization is perceived as a weakness of TOPSIS and other methods based on a distance measure, because the ranking of alternatives depends on the normalization technique and the distance measure. Thus, ROR applied to TOPSIS does not only facilitate the preference elicitation but also solves the problem of non-meaningfulness of TOPSIS. Finally, an instructive example is given to illustrate the proposed method.
3
Publication available in full text mode
Content available

Alternatywne oceny ryzyka systematycznego w modelu CAMP

75%
Trzpiot G.
Studia Ekonomiczne
|
2013
|
vol. 162
9-20
EN
In linear regression model, estimated by last square method, the coefficient of determination gives as an information about ratio of variance of dependence variable describe by chosen in linear relation independence variable. We give the new range of this concept by description the coefficient of determination for chosen robust regression models. We proposed the description of the problem in economic contests, instead that the problem of measurement of systematic risk is a very general issue.
4
Publication available in full text mode
Content available

Preference-Driven Multiobjective Optimization Using Robust Ordinal Regression for Cone Contraction

75%
Kadziński M., Słowiński R.
Multiple Criteria Decision Making
|
2013
|
vol. 8
67-83
EN
We present a new interactive procedure for multiobjective optimization problems (MOO), which involves robust ordinal regression in contraction of the preference cone in the objective space. The most preferred solution is achieved by means of a systematic dialogue with the decision maker (DM) during which (s)he species pairwise comparisons of some non-dominated solutions from a current sample. The origin of the cone is located at a reference point chosen by the DM. It is formed by all directions of isoquants of the achievement scalarizing functions compatible with the pairwise comparisons of non-dominated solutions provided by the DM. The compatibility is assured by robust ordinal regression, i.e. the DM's statements concerning strict or weak preference relations for pairs of compared solutions are represented by all compatible sets of weights of the achievement scalarizing function. In successive iterations, when new pairwise comparisons of solutions are provided, the cone is contracted and gradually focused on a subregion of the Pareto optimal set of greatest interest. The DM is allowed to change the reference point and the set of pairwise comparisons at any stage of the method. Such preference information does not need much cognitive e ort on the part of the DM. The phases of preference elicitation and cone contraction alternate until the DM nds at least one satisfactory solution, or there is no such solution for the current problem setting.
5

Odporność metody najgłębszej regresji na obserwacje nietypowe - analiza porównawcza

63%
Pruska D.
Wiadomości Statystyczne. The Polish Statistician
|
2006
|
issue nr 12
18-25
XX
Obserwacje nietypowe, występujące w zbiorach danych zmniejszają dokładność oszacowań parametrów modeli uzyskanych za pomocą metody najmniejszych kwadratów. Analiza zbiorów danych zawierających obserwacje odstające wymaga stosowania metod odpornych na te obserwacje. W artykule dokonano porównania odporności na obserwacje nietypowe dwóch metod estymacji parametrów funkcji regresji: najgłębszej regresji (MNR) i najmniejszych kwadratów (MNK). Analizie poddano wpływ obserwacji nietypowych na wartości oszacowań parametrów. Przeprowadzone symulacje Monte Carlo we wszystkich rozważanych przypadkach potwierdziły większą odporność na obserwacje nietypowe metody najgłębszej regresji niż metody najmniejszych kwadratów. Porównanie średnich błędów względnych oraz oszacowań parametrów modeli otrzymanych na podstawie metody najmniejszych kwadratów i metody najgłębszej regresji pozwalają stwierdzić, że modele otrzymane za pomocą drugiej metody były lepiej dopasowane do danych, niezależnie od wielkości zakłócenia. (abstrakt oryginalny)
EN
The estimation of regression parameters for data set containing outliers needs the application of robust estimation methods. In the paper the deepest regression method is considered in case of bivarite variables. Some Monte Carlo experiments with outliers are conducted and estimation results for the deepest regression method and ordinary least square method are compared. Experiments confirmed that the deepest regression method is more robust for outliers in the data set than the least square method. (original abstract)
first rewind previous Page / 1 next fast forward last
JavaScript is turned off in your web browser. Turn it on to take full advantage of this site, then refresh the page.