engleski

PROBABILITY DISTRIBUTION SELECTION USING PROMETHEE GDSS METHOD

Rising inequalities between countries are widely investigated. Since they involve diff erent aspects, in this paper they are examined through the distribution of countries’ growth and development, inclusion and intergenerational equity and sustainability. These aspects involve variables that are not normally distributed so diff erent probability density functions (PDFs) have been proposed, tested and compared through literature. This paper compares diff erent PDFs that are considered successful in describing distribution of the variables involved, as Weibull, gamma, lognormal, normal, loglogistic, Pareto, Burr and exponential. They are usually compared using diff erent goodness-of-fi t measures including the log-likelihood, sum of squared errors, sum of absolute errors and chi-square statistics which can sometimes lead to divergent conclusions. This paper adds up to these goodness-of-fi t measures and includes the Kolmogorov-Smirnov, Cramer-von Mises and Anderson-Darling statistics, the mean absolute and squared deviation between theoretical and empirical PDF, the mean absolute and squared deviation between theoretical and empirical cumulative distribution function, AIC and BIC as well as the deviation in skewness and kurtosis. Since diff erent distributions can be considered as alternatives and goodness-of-fi t measures as confl icting criteria, the problem of fi nding the appropriate PDF can be viewed as multiple criteria decision making problem which can be solved using PROMETHEE method. However, diff erent preference parameters, lead to diff erent ranking, so the fi nal decision is obtained using PROMETHEE Group Decision Support System (GDSS) method. This paper therefore contributes to the existing literature on inequalities between countries and statistics in general by proposing a new approach for distribution modelling and selection.

goodness-of-fit, probability density function, PROMETHEE GDSS method.

nije evidentirano