|
A New Goodness-of-Fit Test for a Distribution by the Empirical Characteristic Function
|
M. Towhidi   1, M. Salmanpour |
| 1- , mtowhidi@susc.ac.ir |
|
|
Abstract: (3796 Views) |
Extended Abstract. Suppose n i.i.d. observations, X1, …, Xn, are available from the unknown distribution F(.), goodness-of-fit tests refer to tests such as
H0 : F(x) = F0(x) against H1 : F(x) $neq$ F0(x).
Some nonparametric tests such as the Kolmogorov--Smirnov test, the Cramer-Von Mises test, the Anderson-Darling test and the Watson test have been suggested by comparing empirical distribution, Fn(x), and the known distribution F0(x).
The characteristic function is important in characterizing the probability distribution theoretically. Thus it have been expected that the empirical characteristic function, cn(t), can be used for suggesting a goodness-of-fit test...[To Continue click here] |
|
| Keywords: . characteristic function, consistent test, eigen values, goodness-of-fit test, multivariate central limit theorem, principal components method |
|
|
Full-Text [PDF 403 kb]
(930 Downloads)
|
Type of Study: Research |
Subject:
General
Received: 2016/02/9 | Accepted: 2016/02/9 | Published: 2016/02/9
|
|
|
|
|
|
