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

Results found: 8

first rewind previous Page / 1 next fast forward last

Search results

Search:
in the keywords:  goodness of fit
help Sort By:

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

Wpływ szerokości skali na miary dopasowania modelu ścieżkowego

100%
Skowron Ł.
Econometrics. Ekonometria. Advances in Applied Data Analytics
|
2015
|
issue 2 (48)
23-32
EN
Research problem analysed in the paper concentrates on the impact of the choice of scale on the the goodness of fit of the Structural Equation Model. For that purpose the primary research tool were built and used for conducting two rounds of research. In the first round two groups of respondents were given two different questionaries dealing with the same research problem (same set of research questions) but using different, 5 or 10 point scales. After one week the same groups of respondents were asked again to assess the research problem by using different type of scale. Obtained empirical data were used to build the structural equation models (for 5 and 10 point scales) of the analysed research problem. Thanks to the collected empirical data it was possible to analyze the impact of the choice of scale on internal (AVE, Cronbach’s Alfa, Composite Reliability) and external (R2) goodness of fit of the model.
2
Publication available in full text mode
Content available

Sujatha Distribution and its Applications

75%
Shanker R.
Statistics in Transition new series
|
2016
|
vol. 17
|
issue 3
391-410
EN
In this paper a new one-parameter lifetime distribution named “Sujatha Distribution” with an increasing hazard rate for modelling lifetime data has been suggested. Its first four moments about origin and moments about mean have been obtained and expressions for coefficient of variation, skewness, kurtosis and index of dispersion have been given. Various mathematical and statistical properties of the proposed distribution including its hazard rate function, mean residual life function, stochastic ordering, mean deviations, Bonferroni and Lorenz curves, and stress-strength reliability have been discussed. Estimation of its parameter has been discussed using the method of maximum likelihood and the method of moments. The applications and goodness of fit of the distribution have been discussed with three real lifetime data sets and the fit has been compared with one-parameter lifetime distributions including Akash, Shanker, Lindley and exponential distributions.
3
Publication available in full text mode
Content available

Adekwatność wybranych rozkładów teoretycznych dochodów w zależności od metody aproksymacji

75%
Ostasiewicz K.
Przegląd Statystyczny
|
2013
|
vol. 60
|
issue 4
499-521
PL
Rozkłady dochodów modelowane są za pomocą wielu rozkładów teoretycznych, których parametry wyznaczane są za pomocą różnych metod. Wśród rozkładów tych wymienić można rozkład logarytmiczno-normalny, gamma, czy logarytmiczno-logistyczny. Najczęściej stosowanymi metodami są metoda najmniejszych kwadratów oraz metoda największej wiarygodności. Miarą dobroci dopasowania rozkładu teoretycznego jest zazwyczaj średnie odchylenie kwadratowe lub wartość statystyki chi kwadrat. Tak zdefiniowana jakość dopasowania nie musi jednakże dokładnie przekładać się na jakość oszacowania takich charakterystyk rozkładu, jak wartość średnia czy miary nierówności. Celem pracy jest aproksymacja dochodów różnymi rozkładami teoretycznymi (log-normalny, gamma, log-logistyczny, Daguma i Singh-Maddala) oraz różnymi technikami i porównanie dobroci dopasowania mierzonej jako średnie odchylenie kwadratowe z dokładnością oszacowania kilku charakterystyk liczbowych rozkładu (wartości średniej, odchylenia standardowego, współczynnika Giniego, współczynnika Theila i trzech współczynników Atkinsona), których wartości porównywane są z wartościami dokładnymi, policzonymi na podstawie danych niezgrupowanych.
EN
There are various theoretical distributions which are used as models for the distribu-tion of income. Among them the most commonly used is probably log-normal distribution, but also gamma distribution or log-logistic one. There are also a number of approximation methods of these distributions. The goodness of fit is commonly measured by mean squared deviation or value of chi-square statistics. However, such measures of quality of approxima-tion do not necessarily coincide with accuracy of some distribution characteristics, like ine-quality measures. The aim of this paper is to investigate the goodness of approximation of income distribution, given in the form of frequency distribution, by means of chosen theo-retical distributions, namely, log-normal, gamma, log-logistic, Dagum and Singh-Maddala distribution. The goodness is measured both by mean squared error and deviation of some distribution characteristics. Values calculated on ungrouped data are used as a reference for comparisons.
4
Publication available in full text mode
Content available

Goodness and lack of fit tests to pretest normality when comparing means

75%
FLORES P., PALACIOS M. d. L.
Operations Research and Decisions
|
2024
|
vol. 34
|
issue 1
119-129
EN
Previous studies show that processes related to traditional pretests to prove the perfect fulfillment of assumptions in comparison means tests lead to severe alterations in the overall Type I error probability and power. These problems seem to be overcome when pretests based on an equivalence approach are used. The paper proposes a lack of fit tests based on equivalence to pretest normality on homoscedastic samples with measurable departures from normality. The Type I error probability and power produced by this equivalence pretest are compared with two traditional goodness of fit pretests and with the direct use of the t-Student and Wilcoxon test of means comparison. Furthermore, since the irrelevance limit for the lack of fit test is an arbitrary value, we propose a non-subjective methodology to find it. Results show that this proposed equivalence test controls the overall Type I Error Probability and produces adequate power; therefore, its use is recommended.
5
Publication available in full text mode
Content available

A Two-Parameter Lindley Distribution

75%
Shanker R., Mishra A.
Statistics in Transition new series
|
2013
|
vol. 14
|
issue 1
45-56
EN
A two-parameter Lindley distribution, of which the Lindley distribution (LD) is a particular case, has been introduced. Its moments, failure rate function, mean residual life function and stochastic orderings have been discussed. The maximum likelihood method and the method of moments have been discussed for estimating its parameters. The distribution has been fitted to some data-sets to test its goodness of fit.
6
Content available remote

Measuring the Quality of Multivariate Statistical Models

63%
Trzęsiok M.
Acta Universitatis Lodziensis. Folia Oeconomica
|
2018
|
vol. 6
|
issue 339
99-100
PL
Bardzo ważnym elementem procesu modelowania statystycznego jest etap oceny jakości zbudowanego modelu. W zależności od wykorzystanej metody istnieje wiele różnych podejść do pomiaru jakości modelu. Pomiar ten może skupiać się na dopasowaniu do danych empirycznych albo może przede wszystkim uwzględniać zdolności prognostyczne modelu. Mierniki mogą być absolutne albo względne. Zestaw mierników jakości modelu obejmuje liczną grupę propozycji, z których analityk musi wybrać najodpowiedniejszy do danej sytuacji. W artykule przedstawiono zestawienie mierników jakości modelu oraz sugestię używania innych mierników jakości na etapie wyboru wariantu modelu oraz na etapie oceny jakości modelu końcowego.
EN
Assessing the quality of a statistical model is very important, since it is crucial for the utility of the modelling process’ outcome. There are many different ways of measuring statistical models’ quality. Some of the measures represent a “goodness of fit” approach, some are “prediction ability” orientated. Among them there are absolute and relative measures. It is a researcher’s decision, which model quality measure is the most adequate for the given task. In the paper we present an overview of statistical models’ quality measures and a suggestion of using different ones during the model type selection stage and the stage of assessing the quality of the final model.
7
Publication available in full text mode
Content available

A New Quasi Sujatha Distribution

63%
Shanker R., Shukla K. K.
Statistics in Transition new series
|
2020
|
vol. 21
|
issue 3
53-71
EN
The aim of this paper is to introduce a new quasi Sujatha distribution (NQSD), of which the following are particular cases: the Sujatha distribution devised by Shanker (2016 a), the sizebiased Lindley distribution, and the exponential distribution. Its moments and momentsbased measures are derived and discussed. Statistical properties, including the hazard rate and mean residual life functions, stochastic ordering, mean deviations, Bonferroni and Lorenz curves and stress-strength reliability are also analysed. The method of moments and the method of maximum likelihood estimations is discussed for estimating parameters of the proposed distribution. A numerical example is presented to test its goodness of fit, which is then compared with other one-parameter and two-parameter lifetime distributions.
8
Publication available in full text mode
Content available

Assessment of the influence of dependent variable distribution on selected goodness of fit measures using the example of customer churn model

63%
Migut G.
Econometrics. Ekonometria. Advances in Applied Data Analytics
|
2020
|
issue vol. 24, nr 1
51-70
EN
Classification models enable optimal actions to be taken at every stage of the customer’s lifecycle. A circumstance affecting both the model building process and the assessment of their discriminatory power is the unbalanced distribution of the dichotomous dependent variable. The article focuses on the question of reliable assessment of the goodness of fit. The first part of the article reviews the measures of predictive power and then assesses the impact of the distribution of the dependent variable on the selected measures of goodness of fit. As a result, the high sensitivity of a number of measures such as lift, accuracy (ACC), or F-Score was observed. The sensitivity of MCC and Kappa Cohen’s measurements was also observed. Sensitivity (SENS) and specificity (SPEC), Youden’s index and measures based on ROC curves showed no such sensitivity. The conclusions obtained may allow the avoidance of misjudging the predictive power of models built for both learning and business practice.
PL
Modele klasyfikacyjne umożliwiają podejmowanie optymalnych działań na każdym etapie cyklu życia klienta. Okolicznością wpływającą zarówno na proces budowy modeli, jak i na ocenę ich siły dyskryminacyjnej jest niezbalansowany rozkład dwustanowej zmiennej zależnej. W artykule skoncentrowano się na kwestii wiarygodnej oceny dobroci dopasowania. W pierwszej części artykułu dokonano przeglądu miar siły dyskryminacyjnej, następnie przeprowadzono ocenę wpływu rozkładu zmiennej zależnej na wybrane miary dobroci dopasowania. W wyniku badań zaobserwowano wysoką wrażliwość szeregu miar, takich jak lift, accuracy (ACC) czy F-Score. Zaobserwowano wrażliwość miar MCC oraz Kappa Cohena. Czułość (SENS) oraz specyficzność (SPEC), jak również pochodne miary oparte na krzywej ROC, a także indeks Youdena wykazały brak takiej wrażliwości. Uzyskane wnioski mogą pozwolić na uniknięcie błędnej oceny zdolności predykcyjnej modeli zarówno budowanych na potrzeby nauki, jak i wykorzystywanych w praktyce biznesowej.
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.