• Communications for Statistical Applications and Methods
• /
• 제3권2호
• /
• pp.37-42
• /
• 1996

Maximum likelihood estimation of Cholesky decomposition is considered under normality assumption. It is shown that maximum liklihood estimation gives a Cholesky decomposition of the sample covariance matrix. The joint distribution of the maximum likelihood estimators is derived. The ussual algorithm for a Cholesky decomposition is shown to be equivalent to a maximumlikelihood estimation of a Cholesky root when the underlying distribution is a multivariate normal one.

• 경영과학
• /
• 제19권2호
• /
• pp.155-168
• /
• 2002

In this paper, some efficient implementation techniques for the multifrontal method, which can be used to compute the Cholesky factor of a symmetric positive definite matrix, are presented. In order to use the cache effect in the cache-based computer architecture, a hybrid method for factorizing a frontal matrix is considered. This hybrid method uses the column Cholesky method and the submatrix Cholesky method alternatively. Experiments show that the hybrid method speeds up the performance of the supernodal multifrontal method by 5%～10%, and it is superior to the Cholesky method in some problems with dense columns or large frontal matrices.

• Journal of applied mathematics & informatics
• /
• 제28권3_4호
• /
• pp.913-921
• /
• 2010

The derivative influence measure is adapted to the Cholesky decomposition of a covariance matrix. Formulas for the derivative influence of observations on the Cholesky root and the inverse Cholesky root of a sample covariance matrix are derived. It is easy to implement this influence diagnostic method for practical use. A numerical example is given for illustration.

• 대한산업공학회지
• /
• 제25권3호
• /
• pp.290-297
• /
• 1999

In interior point methods for linear programming, we must solve a linear system with a symmetric positive definite matrix at every iteration, and Cholesky factorization is generally used to solve it. Therefore, if Cholesky factorization is not done successfully, many iterations are needed to find the optimal solution or we can not find it. We studied methods for improving the numerical stability of Cholesky factorization and the accuracy of the solution of the linear system.

• 경영과학
• /
• 제15권1호
• /
• pp.87-95
• /
• 1998

The computational speed of interior point method depends on the speed of Cholesky factorization. The supernodal column Cholesky factorization is a fast method that performs Cholesky factorization of sparse matrices with exploiting computer's characteristics. Three steps are necessary to perform the supernodal column Cholesky factorization : symbolic factorization, creation of the elimination tree, ordering by a post-order of the elimination tree and creation of supernodes. We study performing sequences of these three steps and efficient implementation of them.

• Journal of the Korean Data and Information Science Society
• /
• 제14권4호
• /
• pp.1007-1012
• /
• 2003

A recursive procedure for finding the Cholesky root of the inverse of sample covariance matrix, leading to a direct solution for the inverse of a positive definite matrix, is developed using the likelihood equation for the maximum likelihood estimation of the Cholesky root under normality assumptions. An example of the Hilbert matrix is considered for an illustration of the procedure.

• 한국경영과학회지
• /
• 제28권1호
• /
• pp.51-61
• /
• 2003

In the normal equations approach in which the ordering and factorization phases are separated, the factorization in the augmented system approach is computed dynamically. This means that in the augmented system the numerical factorization should be performed to obtain the non-zero structure of Cholesky factor L. This causes much time to set up the non-zero structure of Cholesky factor L. So, we present a method which can separate the ordering and numerical factorization in the augmented system. Experimental results show that the proposed method reduces the time for obtaining the non-zero structure of Cholesky factor L.

• The Journal of the Acoustical Society of Korea
• /
• 제12권2E호
• /
• pp.47-62
• /
• 1993

SMI 방법은 수치적인 불안정성과 아울러 많은 계산량을 갖는다. 본 논문에서는 역 공분산 행렬의 Cholesky 분할을 이용하여 SMI 방법보다 효율적인 방법을 제안한다. 제안한 방법에서는 적응 빔 형상과 검출이 하나의 구조로 실현되며 이에 피룡한 역 공분산 행렬의 Cholesky factor는 secondary 입력으로부터 GS 프로세서를 이용하여 추정한다. 제안한 구조의 중요한 특징은 공분산 행렬과 Cholesky factor를 직접 구할 필요가 없다는 점이며, 또한 GS 프로세서의 장점을 이용한 systolic 구조를 사용함으로써 효율적인 계산을 수행할 수 있다. 모의 실험을 통하여 제안한 방법의 성능과 SMI 방법의 성능을 서로 비교하였다. 또한 nonhomogeneous 환경에서 동작하기 위한 방법이 제시되었으며, 아울러 계산량이 많은 GS 구조의 단점을 극복하기 위해 lattice-GS 구조를 이용하는 방법을 제안하였다.

• Journal of Communications and Networks
• /
• 제18권1호
• /
• pp.27-39
• /
• 2016

This paper derives and analyzes a novel block fast Fourier transform (FFT) based joint detection algorithm. The paper compares the performance and complexity of the novel block-FFT based joint detector to that of the Cholesky based joint detector and single user detection algorithms. The novel algorithm can operate at chip rate sampling, as well as higher sampling rates. For the performance/complexity analysis, the time division duplex (TDD) mode of a wideband code division multiplex access (WCDMA) is considered. The results indicate that the performance of the fast FFT based joint detector is comparable to that of the Cholesky based joint detector, and much superior to that of single user detection algorithms. On the other hand, the complexity of the fast FFT based joint detector is significantly lower than that of the Cholesky based joint detector and less than that of the single user detection algorithms. For the Cholesky based joint detector, the approximate Cholesky decomposition is applied. Moreover, the novel method can also be applied to any generic multiple-input-multiple-output (MIMO) system.

• 경영과학
• /
• 제14권2호
• /
• pp.69-79
• /
• 1997

The computational speed of interior point method of linear programming depends on the speed of Cholesky factorization. If the coefficient matrix A has dense columns then the matrix A.THETA. $A^{T}$ becomes a dense matrix. This causes Cholesky factorization to be slow. We study an efficient implementation method of the dense column splitting among dense column resolving technique and analyze the relation between dense column splitting and order methods to improve the sparsity of Cholesky factoror.