Testing hypotheses for multivariate normal distribution with fuzzy random variables

Author(s):  
Gholamreza Hesamian ◽  
Mohamad Ghasem Akbari
1971 ◽  
Vol 20 (4) ◽  
pp. 135-142
Author(s):  
K. C. Chanda

Summary The purpose of this article is to investigate the ‘large sample’ properties of sample quantiles when we assume that the basic random variables are exchangeable (ref. Loève (1960) p. 365). It is shown that under different conditions (to be specified below) on the nature of these exchangeable random variables the distribution of the sample quantile Xr : n where Xr : n is the rth order statistic for the first n exchangeable random variables [Formula: see text] tends, as n → ∞, to different nondegenerate forms. As an example, the special case of random variables with equicor-related multivariate normal distribution is discussed.


Mathematics ◽  
2021 ◽  
Vol 9 (4) ◽  
pp. 438
Author(s):  
Viliam Ďuriš ◽  
Renáta Bartková ◽  
Anna Tirpáková

The probability theory using fuzzy random variables has applications in several scientific disciplines. These are mainly technical in scope, such as in the automotive industry and in consumer electronics, for example, in washing machines, televisions, and microwaves. The theory is gradually entering the domain of finance where people work with incomplete data. We often find that events in the financial markets cannot be described precisely, and this is where we can use fuzzy random variables. By proving the validity of the theorem on extreme values of fuzzy quantum space in our article, we see possible applications for estimating financial risks with incomplete data.


1986 ◽  
Vol 114 (2) ◽  
pp. 409-422 ◽  
Author(s):  
Madan L Puri ◽  
Dan A Ralescu

Sign in / Sign up

Export Citation Format

Share Document