Beruflich Dokumente
Kultur Dokumente
Page | 7
http://japmnt.com/
(JPMNT) Journal of Process Management – New Technologies,
Vol. 1, No.1, 2013.
It should be noted that the notion of functions. Defining the operation called
probability does not enter into the measure Superimposition of Sets and using the
theoretic definition of a random variable Glivenko-Cantelli Theorem (Loeve, 1977) on
(Rohatgi and Saleh, 2001, pages 41 – 43). Order Statistic, Baruah (2010a, 2010b, 2011b,
When a variable is probabilistic, it has to be 2011c, 2011d, 2012) has established the
random by definition although when a following result which we shall state as a
variable is random, it need not be theorem that uncovers the missing link
probabilistic. Accordingly, all results of the between fuzziness and randomness, which
classical theory of probability are was being searched for by the workers in
automatically applicable to a random variable fuzziness since 1965.
defined in the measure theoretic sense. In the Theorem 1: For a normal fuzzy
term ‘fuzzy randomness’, as available in the number
literature, the word ‘randomness’ has been N = [α, β, γ]
taken as equivalent to the word with membership function
‘probabilistic’. 1 x , x
In what, follows, we shall discuss in N x 2 x , x
short Baruah’s Randomness – Fuzziness 0,
otherwise
Consistency Principle, and we shall state the
such that
axiom defining the complement of a fuzzy
Ψ1 (α) = Ψ2 (γ) = 0,
set. Thereafter we shall discuss how exactly
Ψ1 (β) = Ψ2 (β) = 1,
to define fuzzy randomness. Finally, we shall
Ψ1(x) is the distribution function of a
discuss the matters with reference to testing a
random variable defined in the interval [α, β],
fuzzy hypothesis in the case of the Student’s
and Ψ2(x) is the complementary distribution
t-test.
function of another random variable defined
in the interval [β, γ].
1. The Randomness – Fuzziness
It needs to be mentioned at this point
Consistency Principle
that the Glivenko – Cantelli theorem on
convergence of empirical probability
A normal fuzzy number N = [α, β, γ] is
distributions can actually be seen as the
an interval around the real number β with the
backbone of mathematical statistics. This
elements in the interval being partially
theorem is about probability distribution
present. Partial presence of an element in a
functions, and therefore it will be applicable
fuzzy set is defined by the membership
for distribution functions of random variables
function. A normal fuzzy number N = [α, β, γ]
with randomness defined in the measure
is associated with a membership function μN
theoretic sense as well (Baruah, 2011b, 2012).
(x), where
As we have mentioned earlier, in the measure
1 x , x
theoretic sense, if a variable is probabilistic, it
N x 2 x , x
0, has to be necessarily random, although when
otherwise
a variable is random, it does not have to be
Here Ψ1(x) is continuous and non- probabilistic.
decreasing in the interval [α, β], and Ψ2(x) is It is known that a distribution function
continuous and non-increasing in the interval of a random variable is non-decreasing, and
[β, γ], where that a complementary distribution function of
Ψ1 (α) = Ψ2 (γ) = 0, a random variable is non-increasing. The
Ψ1 (β) = Ψ2 (β) = 1. functions are continuous and differentiable.
In the Dubois-Prade nomenclature, Differentiation of Ψ1(x) and (1 – Ψ2(x)) would
Ψ1(x) is called the Left Reference Function, give two density functions. This means, one
and Ψ2(x) is called the Right Reference needs two laws of randomness, one in the
Function of the normal fuzzy number. interval [α, β] and the other in [β, γ], to
Construction of a normal fuzzy number would construct a normal fuzzy number [α, β, γ].
depend on construction of these two reference
Page | 8
http://japmnt.com/
(JPMNT) Journal of Process Management – New Technologies,
Vol. 1, No.1, 2013.
x dx 1 .
1
4. Student’s – t Test with Fuzzy Data
In the same way, the equivalence of Assume that a random sample of five
the definitions of the Dubois-Prade right students has been collected from a normal
reference function Ψ2(x), β ≤ x≤ γ, and a population and their heights are measured.
complementary distribution function, gives us The heights of the five students are 63, 67, 70,
d
1 2 x 2 x , say, 71 and 73 inches. Let us assume that mean
dx height of the population is 66 inches.
where Now, we want to test whether the data
are consistent with the assumption of a mean
x dx 1.
2 height of 66 inches in the population, i.e., H0:
μ=66, against the alternative hypothesis, H1:
Now, according to Baruah’s μ≠66.
Randomness – Fuzziness Consistency Under H0, the test statistic is given by
Principle, a triangular fuzzy number of the
x 0
type t 2 ,
X i xi , xi , xi s / n 1
with membership function which follows the Student’s – t
probability distribution with (n-1) degrees of
x xi
, if xi x xi , freedom, where x is the sample mean and
s 2 is the sample variance.
x x
X i x i , if xi x xi , Here, the calculated value of t is
0, otherwise 1.4364 which is less than the tabulated value
of t i.e. 2.78 at 5% probability level of
Page | 10
http://japmnt.com/
(JPMNT) Journal of Process Management – New Technologies,
Vol. 1, No.1, 2013.
Page | 11
http://japmnt.com/
(JPMNT) Journal of Process Management – New Technologies,
Vol. 1, No.1, 2013.
therefore the probability that the fuzzy null 3. Baruah, H. K. (2010a), The randomness –
hypothesis that the sample has come from a fuzziness consistency principle, International Journal
of Energy, Information and Communications, Vol. 1,
population with fuzzy mean height 66 inches Issue 1, 37 – 48.
would have to be rejected at 5% probability 4. Baruah, H. K. (2010b), The Randomness-
level of significance. In other words, when a Fuzziness Consistency Principle, Invited Article,
non-rejectable hypothesis is fuzzified, there International Congress of Mathematics, Proceedings of
will still be a probability that the fuzzy the Satellite International Conference on Probability
and Statistics, Sambalpur University, India.
hypothesis would actually be found 5. Baruah, H. K. (2011a), Towards Forming a
rejectable. In the same way, if a rejectable Field of Fuzzy Sets, International Journal of Energy,
hypothesis is fuzzified, there would still be a Information and Communications, Vol. 2, Issue 1, 16 –
probability that the fuzzy hypothesis would be 20.
found non-rejectable, the probability of non- 6. Baruah, H. K. (2011b), Construction of the
membership function of a fuzzy number, ICIC Express
rejection being decided by the left reference Letters, Vol. 5, Issue 2, 545-549.
function this time. 7. Baruah, H. K. (2011c), The theory of fuzzy
sets: beliefs and realities, International Journal of
5. Conclusions Energy Information and Communications, Vol. 2, No.
2, 1-22.
8. Baruah, H. K. (2011d), In search of the root of
Partial presence of an element in a fuzziness: the measure theoretic meaning of partial
normal fuzzy number can be expressed in presence, Annals of Fuzzy Mathematics and
terms of two laws of randomness. The Informatics, Vol. 2, No. 1, 57-68.
membership function of a normal fuzzy 9. Baruah, H. K. (2012), Construction of normal
number [α, β, γ] is actually a distribution fuzzy numbers using the mathematics of partial
presence, Journal of Modern Mathematics Frontier,
function in [α, β] and a complementary Vol. 1, No. 1, 9-15.
distribution function in [β, γ]. Based on this 10. Buckley, J. J. (2003), Uncertain probabilities
principle, the exact definition of fuzzy III: the continuous case, Soft Computing – A Fusion of
randomness has been forwarded in this article. Foundations, Methodologies and Applications, Vol. 8,
Further, the complement of a fuzzy set should No. 3, 200 – 206.
11. Buckley, J. J. and Eslami, E. (2003),
be redefined. In testing of fuzzy hypothesis, Uncertain probabilities I: the discrete case, Soft
we deal with the alternative hypothesis which Computing – A Fusion of Foundations, Methodologies
is the complement of the fuzzy null and Applications, Vo. 7, No. 8, 500 – 505.
hypothesis. One should note that for the 12. Buckley, J. J. and Eslami, E. (2004),
complement of a normal fuzzy set, fuzzy Uncertain probabilities II: the continuous case, Soft
Computing – A Fusion of Foundations, Methodologies
membership function and fuzzy membership and Applications, Vol. 8, No. 3, 193 – 199.
value are two different things. We have 13. Das, D., Mahanta S., Chutia, R. and Baruah,
shown that when a non-rejectable hypothesis H. K. (2013), Construction of normal fuzzy numbers:
is fuzzified, there would still be a probability case studies with Indian stock exchange data, Annals of
that the fuzzy hypothesis would actually be Fuzzy Mathematics and Informatics (in press).
14. Goswami, P. and Baruah, H. K. (2008a),
found rejectable. In the same way, if a Fuzzy Discrete Distributions: the Binomial Case,
rejectable hypothesis is fuzzified, there would Journal of Fuzzy Mathematics, Vol. 16, No. 3, 671 –
still be a probability that the fuzzy hypothesis 676.
would be found non-rejectable. 15. Goswami, P. and Baruah, H. K. (2008b), The
Fuzzy ARIMA (1, 1) Process, Vol. 16, No. 3, 721 –
729.
Reference 16. Goswami, P. and Baruah, H. K., (2007),
Fuzzy Time Series Analysis, Journal of Fuzzy
1. Baruah, H. K. (1999a), Set Superimposition Mathematics, Vol. 15, No. 3, 513 -523.
and Its Application to the Theory of Fuzzy Sets, 17. Goswami, P., Dutta, P. and Baruah, H. K.
Journal of the Assam Science Society, Vol. 40, Nos. 1 (1997), The Latin Square Design Using Fuzzy Data,
& 2, 25 – 31. Journal of Fuzzy Mathematics, Vol. 5, No. 4, 767 -779.
2. Baruah, H. K. (1999b), Fuzzy Membership 18. Loeve, M. (1977), Probability Theory,
with respect to a Reference Function, Journal of the Springer Verlag, New York.
Assam Science Society, Vol. 40, No. 3, 65 – 73. 19. Rohatgi V. K. and Saleh A. K. E. (2001), An
Introduction to Probability and Statistics, Second
Page | 12
http://japmnt.com/
(JPMNT) Journal of Process Management – New Technologies,
Vol. 1, No.1, 2013.
Edition, Wiley Series in Probability and Statistics, John 22. Talukdar, R. and Baruah, H. K. (2010b), A
Wiley & Sons (Asia) Pte Ltd., Singapore. Sequential Variance Ratio Test with Fuzzy
20. Talukdar, R. and Baruah, H. K. (2007), Observations, Advances in Fuzzy Mathematics, Vol. 5,
Sequential Probability Ratio Test with Fuzzy No. 3, 301-309.
Observations, Journal of Fuzzy Mathematics, Vol. 15, 23. Talukdar, R. and Baruah, H. K. (2010c), A
No. 3, 627 – 643. Two Sample Sequential t-Test with Fuzzy
21. Talukdar, R. and Baruah, H. K. (2010a), A Observations, Applied Mathematical Sciences, Vol. 4,
Fuzzy Sequential Decision Procedure Applied to No. 68, 3361 – 3374.
Completely Randomized Design of Experiments, 24. Talukdar, R. and Baruah, H. K. (2011), A
International Journal of Contemporary Mathematical Sequential Multi-Decision Procedure with Fuzzy
Sciences, Vol. 5, No. 26, 1283 – 1302. Observations, Journal of Fuzzy Mathematics, Vol. 19,
No. 1, 125 – 140.
Page | 13
http://japmnt.com/