Questions on the various algorithms used in linear algebra computations (matrix computations).

learn more… | top users | synonyms

0
votes
0answers
96 views

QR fatorization for tridiagonal matrices

Let $$A = \left[\begin{array}{rrrr} \delta_1&\gamma_2 & &0 \\ \gamma_2&\delta_2 &\ddots & \\ &\ddots &\ddots &\gamma_n \\ 0 & &\gamma_n ...
0
votes
0answers
21 views

the rank of QR decomposition

I saw this in a paper, where one has a QR decomposition $C=QR$ ($C\in R^{m\times r}$, $Q\in R^{m\times r}$ is column orthogonal, $R\in R^{r\times r}$, $m>r$). However, under the condition that the ...
0
votes
0answers
34 views

Practical examples/implementation details for Gauss-Seidel method

I'm having a presentation on Gauss-Seidel iterative method, and although it isn't mandatory , I would like to have some practical examples for this method (a system of linear equations with n>=1000, ...
0
votes
0answers
16 views

Iterated Schur complement for block matrices

Suppose you have got a symmetric block matrix $A = \begin{pmatrix} A_{1,1} & \dots & A_{1,n} \\ \vdots & & \vdots \\ A_{n,1} & \dots & A_{n,n} \end{pmatrix}$ Suppose that ...
0
votes
0answers
37 views

How to fit a stochastic matrix to given data.?

Given a data sequence of noisy observations of a 3-state Markov chain $X$ -- $y_1$,$y_2$,...$y_n$, with two transition matrices $A_1$ and $A_2$ corresponding to different regions (**) in the (unit) ...
0
votes
0answers
11 views

Powers of matrices via the generalised Lanczos process

At each iterative step of the generalised Lanczos process for the pair of matrices (A,B), we obtain the following factorisation: $$ A Q_k = B Q_{k+1} \widehat{T}_k, $$ where $Q_k^T B Q_k = I_k$ and ...
0
votes
0answers
42 views

Which of the following fixed point iterations will converge?

Which of the following fixed point iterations will converge? Why? Give the rate of convergence. (a) $x_{n+1} = \cos x_n$ (b) $x_{n+1} = \sin x_n$ (c) $x_{n+1} = \tan x_n$ For $10$ bonus ...
0
votes
0answers
18 views

Mixed Lognormal Model Calibration

Any ideas as to how to calibrate a mixed lognormal volatility model (Brigo and Mercurio 2002) for arbitrary N < 10? The paper seems vague with respect to implementation.
0
votes
0answers
13 views

LU Decopmostions with block

So both $A_{11}$ and $\hat{A_{22}}$ have $LU$ decompositions say $A_{11}=L_{1}U_{1}$ and $\hat{A_{22}}=L_{2}U_{2}$. Show that $ \begin{bmatrix} A_{11} & A_{12} \\ A_{21} & A_{22} ...
0
votes
0answers
35 views

Generalized SVD and weighted SVD

I've the following question: How should I select the $A$,$B$ matrices in the generalized singular value decomposition (GSVD) such that it solves the weighted version of the generalized singular value ...
0
votes
0answers
40 views

Finding the closest low rank correlation matrix?

I am looking to find the rank 3 correlation matrix approximation of a rank $n-1$ correlation matrix. This best approximation can be more clearly defined as the closest correlation matrix with rank 3 ...
0
votes
0answers
22 views

Linear System with non zeros count constraint

I trying to solve a simple linear system: $Ax=b$ But with constraints like: $\sum{x_i}=S$, Usually S = 1. $L \le x \le U$, Lower & Upper bounds (usually $0 \le x \le 1$) And "Maximum count of ...
0
votes
0answers
15 views

Does the Conjugate Gradient Method provide an eigenvalue estimate?

Suppose that we apply a Krylov subspace method to the linear system $A x = b$. For example, if $A$ is symmetric positive-definite, then the Conjugate Gradient method may be used. I remember that the ...
0
votes
0answers
30 views

Search Direction in Conjugate Gradient

Could you help me with a Conjugate Gradient question? In using CG to solve $Ax = b$, why is the search direction $p_{k+1}$ in CG chosen as a linear combination of the residual $r_k$ and previous ...
0
votes
0answers
14 views

choice of iterative linear system method

while implementing an unconstrained optimization problem, using Newton's method, I am faced with a Hessian matrix that is very large (10^8 by 10^8) but very, very sparse - Non zero elements along the ...
0
votes
0answers
55 views

Induced matrix p-norm

Let $\|\cdot\|_p$ denote the $p$ norm $(p≥1)$ defined for every vector $x=(x_1,x_2,\ldots,x_n)^t\in\mathbb C^n$ by $\|x\|_p=(\sum|x_j|^p)^{1/p}$ and let $|||\cdot|||_p$ denote the matrix norm defined ...
0
votes
0answers
74 views

update cholesky factorization

I need to compute cholesky(H'*H) where H is a big sparse rectangular matrix. After that H is modified by adding several lines. That is Hn = [H ; line_1 ; ... ; line_n] in Matlab. How can I recompute ...
0
votes
0answers
19 views

order of convergence for approximations

Let $u \in L^{2}(0,1)$ and $0 < x_{1}< x_{2}<... < x_{n} = 1$, where x$_{k}$ = k$\cdot$h, n$\cdot$h = 1, a partition of the interval [0,1]. Define I$_{k}$(x) = 1 if x $\in$ [x$_{k}$, ...
0
votes
0answers
19 views

Location and perturbation of eigenvalues

This is a problem from Horn and Johnson's Matrix Analysis. I'm having trouble showing the bolded parts in the following paragraphs. In fact, I don't really understand what the sentences mean. I would ...
0
votes
0answers
28 views

A problem on Gersgorin cirle passing through the eigenvalue of an absolute matrix

I'm having trouble solving the following problem. I think I need to show that the matrix $D^{-1}|A|D$ has property SC, but I can't come up with a way to show it. I would really appreciate any ...
0
votes
0answers
32 views

Is the following matrix Upper Hessenberg?

Does $$ A = \begin{pmatrix} 1 & 1 \\ -1 & -1 \end{pmatrix}$$ properly satisfy the definition of upper Hessenberg?
0
votes
0answers
251 views

Applying Central Difference (Finite Difference Method) in MATLAB

I was given a rather complicated few problems to solve in MATLAB using the central difference method, and I'd like some help figuring out how to translate this into code. The goal is to discretize ...
0
votes
0answers
35 views

Show that Newton’s Method is well-defined for all k and converges to 0 for $x_0>0$

Let $f : R → R$ with $f$ twice continuously differentiable, $\gamma > f''(x)>\delta, f(0)=0,f'(x)>\rho $ for $x ≥ 0$. Show that for any $x_0 > 0$ that Newton’s Method is well-defined for ...
0
votes
0answers
21 views

Jacobi Iteration with Shift

The question is to solve a linear system using Jacobi iterations with a shift of mu = 5. My code converges very quickly, but it does not yield the results that MATLAB gives with the backslash ...
0
votes
0answers
17 views

Convergence of recursive application of finite-difference operator to $C^{\infty}$ functions

Let $f\colon \mathbb{R}\to \mathbb{R}$ be an arbitrary smooth function (whose extension to a complex differentiable function is entire, if it matters). Let $\mathbf{D}_{h}$ be a finite difference ...
0
votes
0answers
10 views

Get stuck with some statements of convergence rate of the iteration method from “Iterative methods for sparse linear systems (2nd edition) ”

Here are the statements I get from the book and the two highlight parts are what I can not understand well. The questions are: Why we can conclude that $\rho=\rho(G)$ from"the above analysis"? ...
0
votes
0answers
24 views

Interpolation question about polinomials?

Let $f(x)=x^n$. show that for each $n$ distinct point $x_0,... x_n$ we have $f[x_0,...x_n]=0$ and $f[x_0,...,x_{n-1}]=\sum_{i=0}^{n-1}x_i$. also show that if $x_{i+1}-x_i$ is fixed for each $i\ge 0$ ...
0
votes
0answers
14 views

Applying perturbed matrix to unperturbed eigenvector

Suppose we've got a matrix $P$ and a perturbed version $\hat{P}=P+E.$ Given that $v$ is an eigenvector of $P$ with $Pv=0,$ I'd like to get as sharp a bound as possible on $\hat{P}v$ (in terms of ...
0
votes
0answers
146 views

Matlab project - Jacobi method for tridiagonal matrices…

I have to do a project in Matlab to my University and I don't quite understand what I should do. I was given script that solves systems of equations with Jacobi's method with given tolerance and ...
0
votes
0answers
42 views

QR Algorithm without Shifts (Trefethen and Bau)

A real symmetric matrix $A$ has eigenvalue 1 of multiplicity 8, while all the rest of the eigenvalues are $\leq 0.1$ in absolute value. Describe an algorithm for finding an orthonormal basis of the ...
0
votes
0answers
59 views

Strictly diagonally dominant matrix -LU factorization

Let $A\in\mathbb{C^{n\times n}}$ be strictly diagonally dominant. I want to show that the LU factorizations with and without partial pivoting are the same for these matrices. For start, I created ...
0
votes
0answers
24 views

Pairing Two Point Clouds

So I have two point clouds $X$ and $Y$ each with $N$ points in the familiar $\mathbb{R}^3$ euclidian 3D space. I then have an inter-point distance $d(\vec x_i,\vec y_j)$ which is zero if $\vec x_i$ is ...
0
votes
0answers
17 views

Relative Error with Respect to Frobenius Norm

I'm look at this tiny book called "Deblurring Images: Matrices, Spectra, and Filtering" by Hansen, Nagy, O'Leary. This is a self study, but I believe my question is broad enough so that it can be of ...
0
votes
0answers
14 views

Dimensionality Reduction

Let $X\in\mathbb{R}^{100\times 100}$ matrix and let its eigenvector and eigenvalues be $X_{vec}$ and $X_{val}$ respectively. If the rank of $X$ is $5$, then is it possible to approximate $X$ with ...
0
votes
0answers
39 views

Condition number perturbation

I have a matrix of the form $\tilde{H} = H + i A A^\dagger$. It is known that $H$ is hermitian and that $\tilde{H}$ is invertible and $A A^\dagger$ has a kernel of dimension $\geq 1$. I want to study ...
0
votes
0answers
32 views

Householder reflector sign error

I am studying Householder reflectors from Trefethen and Bau but am having trouble creating a simple example for it. I am given the equation for the vector v that the Householder reflector H is based ...
0
votes
0answers
30 views

Computing a few eigenvalues of large sparse nonsymmetric matrix without LU factorisation

I'm trying to find a few targeted eigenvalues of a large sparse (N=1e6,nnz=4e6) non-symmetric matrix. Currently I'm using MATLAB's 'eigs' function with the 'sigma' option and this uses the Shifted ...
0
votes
0answers
29 views

Recasting Ax=b to use SOR method

Just need some guidance as to how to recast the matrix equation equation $Ax = b$ so that I can produce an iterative matrix to perform Succesive Over Relaxation on. These matrices are $n x n$ I ...
0
votes
0answers
17 views

Properties of specific case of the gradient descent method

Consider the gradient descent method for the system $Ax=b$ where, $A=\begin{pmatrix}1 & 0 \\0 & a\end{pmatrix}, b=\begin{pmatrix}0 \\ 0\end{pmatrix}$ and the initial vector ...
0
votes
0answers
37 views

matrix diagonalization without eigen decomposition, what other ways available?

I have a matrix, $A$ (it may be symmetric or asymmetric). I need to have a diagonal matrix without eigenvalue decomposition, please suggest what others ways are possible? Any new idea would be much ...
0
votes
0answers
25 views

computing leftmost eigenpair of positive-definite matrix

Let $A$ be an $n\times n$ real symmetric positive-definite matrix. Assume that $n$ is large and that $A$ is dense (i.e. it is not sparse). Question: What is the state-of-the-art algorithmically for ...
0
votes
0answers
45 views

Test for powers method

I have been told that for a normal matrix $A$, the powers method (i.e. computing the succession of Rayleigh quotients for a succession of vectors $z_k=A\cdot z_{k-1}$) can use the following stop ...
0
votes
0answers
38 views

Are there any sure-fire methods for correctly arranging matricies for Gaussian Elimination?

I am attempting to make a Gaussian Elimination solver for systems of linear equations that contain less than 100 equations. I have roughed out a method for creating and filling in the diagonal of a ...
0
votes
0answers
30 views

Stuck on condition number derivation of the perturbed equation $(A + \Delta)\tilde{x} = b + \delta_b$

I've almost got what I want. We start with $Ax = b $ and $(A + \Delta)\tilde{x} = b + \delta_b$. What I have then is \begin{align*} \tilde{x} - x &= -A^{-1}\Delta\tilde{x} + A^{-1}\delta_b \\ ...
0
votes
0answers
46 views

Integration of ODE equation in Matlab / Octave

I have a system of 8 ODE's where the initial conditions are in matrix form. $\frac{dT}{dS} = H T$ where T at the initial state is the identity matrix. $T(a) = I$ H is a constant 8x8 matrix T is ...
0
votes
0answers
51 views

Is the Hessenberg form of a matrix unique?

I have to calculate the Hessenberg form of an matrix using householder reflectors. For real Matrices I get the same result as the 'hess()' function in Matlab, but for Complex Matrices I keep getting ...
0
votes
0answers
75 views

Alternative to the Gram-Schmidt Procedure for Orthogonalization

I was wondering if there is an alternative to the Gram-Schmidt procedure, which instead of being a successive orthogonalization scheme, would be non-successive (simultaneous)? In other words, is there ...
0
votes
0answers
94 views

Proof of theorem about iterative methods

How do I prove that if $A$ is a tridiagonal (or block tridiagonal) matrix then the corresponding $P_J$ and $P_G$ iteration matrices for the Jacobi and Gauss-Seidel methods satisfy that if $\lambda$ is ...
0
votes
0answers
28 views

How is this a substitution? Linear algebra transformation matrix misunderstanding

I found the following matrix equation in '3D Surveillance System Using Multiple Cameras', (authors: Ajay Kumar Mishra, Bingbing Ni, Stefan Winkler, Ashraf Kassima) (link here): I don't follow the ...
0
votes
0answers
37 views

Nearest points / residuals on a total least squares parabola

Consider fitting a parabola $y = a + bx + cx^2$ to 2d data $X_i, Y_i$ with noise in both X and Y, using the the singular value decomposition as in Total_least_squares (TLS): $\qquad X = [ 1\ \ Xdata\ ...