Search

Follow-up experiment is widely applied to various fields such as science and engineering, since it is an indispensable strategy, especially when some additional resources or information become available after the initial design of experiment is carried out. Moreover, some extra factors may be added in the follow-up experiment. One may augment the number of runs and/or factors for the purpose of application. In this paper, the issue of the uniform row augmented designs and column augmented designs with mixed two- and three-level is investigated. The uniformity of augmented designs is discussed under the Lee discrepancy, some lower bounds of Lee discrepancy for the augmented designs are obtained. The construction algorithm of the uniform augmented designs is given. Some numerical examples indicate that uniform augmented designs can be constructed with high efficiency. [ABSTRACT FROM AUTHOR]

Based on the multiplicative update rule of nonnegative matrix factorization (NMF), we add the nonnegative constraint of endmember matrix to the objective function of iterative constrained endmembers (ICE), and propose a simple multiplicative update rule for ICE, named NMF-ICE. Our method avoids the use of quadratic programming method in ICE, therefore can greatly improve the computational efficiency. Experiments using both simulated and real hyperspectral data show that NMF-ICE is effective in endmember generation. [ABSTRACT FROM AUTHOR]

We consider an exact factor model with integrated factors and propose an LM-type test for unit roots in the idiosyncratic component. We show that, for a fixed number of panel individuals (N) and when the number of time points (T) tends to infinity, the limiting distribution of the LM-type statistic is a weighted sum of independent Chi-square variables with one degree of freedom, and whenTtends to infinity followed byNtending to infinity, the limiting distribution is standard normal. The results should contribute to the challenging task of deriving likelihood-based unit-root tests in dynamic factor models. [ABSTRACT FROM AUTHOR]

In this article, we consider experimental situations in which m two-lever factors are to be studied using a two-level main effects plan involving n runs which are partitioned into b blocks of size k = n/b. For the casen ≡ 1(mod 4) and b ≤ k, we derive some new methods of constructing E-optimal designswhich tend to be highly efficient under other optimality criteria as well. [ABSTRACT FROM AUTHOR]

The classical chi-square test for independence cannot convey additional information and the degree of the association of two factors in two-way contingency table. Besides, measures of association by contingency coefficient need to be used with care, because the association measures depend on the number of rows r and the number of columns c. This article proposes a multistage chi-square test to measure the degree of the association between the two factors in two-way contingency table. We also give simulation and real examples to assess the performance of the proposed method. The results show that our proposed method can effectively investigate the degree of association of two factors in two-way contingency table. [ABSTRACT FROM AUTHOR]

In this paper, we define the generalized conditional transform with respect to the Gaussian process. We then establish the integration formulas for the generalized conditional transform with respect to the Gaussian process in terms of the generalized conditional ⋄-product and the first variation. Also, we show that the generalized conditional transform with respect to the Gaussian process for the first variation ofFcan be expressed in terms of the ordinary function space integral ofFmultiplied by a linear factor. [ABSTRACT FROM AUTHOR]

The median absolute deviation is widely used as a robust measure of scale. In this paper, computer simulations were used to calculate finite-sample correction factors for sample sizes to n = 100. Prediction equations for these correction factors are also provided and are shown to eliminate systematic biases between even and odd values of n. [ABSTRACT FROM PUBLISHER]