Questions on numerical analysis/numerical methods; methods for approximately solving various problems that often do not admit exact solutions.

learn more… | top users | synonyms (2)

0
votes
1answer
12 views

order of convergence PDE

If I have the generic PDE \begin{equation} u_t + u_x = f, \end{equation} approximated with a first order in time and a second order in space numerical scheme, how can I show that the solution ...
1
vote
1answer
18 views

ODE, Picard approximation of a second order equation: How do I make sure that this is correct.

I have the following problem: $$\ddot{x} + \dot{x}^2-2x=0$$ and I.V are: $x(0)=1 \qquad$ $\dot{x}(0) = 0$. and I need to find two first "Picard" approximations. I first arranged it in the form ...
2
votes
0answers
32 views

Fourier transform of periodic function

Is it possible to Fourier transform a periodic function f(x) = f(x+L) with period L, numerically, only over the range x = 0 to L and use periodic boundary conditions to enforce the periodicity of the ...
1
vote
0answers
19 views

Adi-method for Diffusion-reaction equation in 2d

i'm trying to solve this pde using an adi-method (alternating-direction-implicit). $\frac{d f}{d t}=D\nabla^2_{x,y} f+Q(x,y)f+C$ After discretizing, the equation looks like this. Implicit in ...
0
votes
1answer
20 views

$O(\text{polylog}(1/\epsilon))$-time Algorithm for Numerical Integration to Within Additive $\epsilon$?

I'm trying to approximate a 1D definite integral to within an additive $\epsilon$ for a given $\epsilon$. I was wondering whether there is an $O(\text{polylog}(1/\epsilon))$-time algorithm for this. ...
0
votes
0answers
20 views

Name of numerical methods for second-order differential equation

Numerical methods that try to solve first-order differential equations of the form: $$ \frac{\partial}{\partial t} y = f(y,t) $$ are often Runge-Kutta methods, and there is a whole family of ...
2
votes
0answers
18 views

Fast Way to Compute DFT with index summation subject to a constraint

I really appreciate if anyone can help me with this problem. Problem: Let $W_n=e^{\frac{2\pi i}{N}}$ which is the $N$th root of unity. The backward Discrete Fourier Transform of a complex vector ...
0
votes
0answers
17 views

Ruling span derivation?

I have recently read a paper about the ruling span for electrical wires and they have an approximation that looks like it can be derived with mathematical intuition only. I'd like to find a derivation ...
-1
votes
1answer
19 views

Finding approximation of root

AS for newton approximation of the reciprocal of the square root of 5. Does the function f(x)=1/x - 5^1/2 apply for newton method
-3
votes
0answers
17 views

numerical methods (Numerical Differentiation) [on hold]

How will we choose the value of "q" in Richardson Extrapolation. "q" can't take on the values "0", "1" and "-1" because of the presence of "q((q^2)-1)" in the denominator of the formula.
2
votes
3answers
61 views

How do i solve this equation ${\mathbb{R}}$: $3 \sin^3x+2 \cos^3x=2 \sin x+\cos x$?

How do I solve this equation ${\mathbb{R}}$: $3 \sin^3x+2 \cos^3x=2 \sin x+\cos x $? Note : I have tried using trigonometric transformation but it seems very complicated to get the result .. may ...
1
vote
0answers
17 views

Direct computation of $\operatorname{log}(\operatorname{cdf})$ for a normal distribution

This question is linked to the normal distribution for a random variable. The probability density function (pdf) is expressed as: \begin{equation} \frac{1}{\sigma\sqrt{2\pi}}\, e^{-\frac{(x - ...
0
votes
1answer
53 views

Determine the roots of equation if possible

How to determine the roots of equation using numerical methods? I have this particular equation: $$\arctan(e^x)=\ln \left(\sqrt{\frac{e^{2x}}{e^{2x}+1}}\right)$$ In my solution I have that this ...
1
vote
0answers
38 views

How to take derivative of $F(u)=\sum_{i=1}^{N} \int f^2(x) u_i^q(x) dx $

I have to find the derivative of a function. Could you help me to find it $$F(u)=\sum_{i=1}^{N} \int_{\Omega} f^2(x) u_i^q(x) dx $$ where $q \ge 1$, $f(x): \Omega \to R$, $u_i$ is membership ...
1
vote
1answer
32 views

Underdetermined vs Overdetermined Problem

I'm trying to create a model which is of the form $$y = (a_0 + a_1l)[b_0+\sum_{m=1}^M b_m\cos(mx-\alpha_m)] [c_0 +\sum_{n=1}^N c_n\cos(nz-\beta_n)]$$ In the above system, $l$,$x$ and $z$ are ...
0
votes
0answers
16 views

Implicit numerical integration: error bound

Suppose I'm solving this equation numerically with a time step $h$: $$x''(t) = f(x)$$ Discretizing it and using implicit integration: $$x^{n+1} - 2x^n + x^{n-1} = h^2f( x^{n+1})$$ $x^{n-1}$ and ...
1
vote
1answer
24 views

Error bound of midpoint rules with unbounded second derivative

It is well known that error bound of midpoint rule for function $f[a,b]$ is given by $$ E\leq K\frac{(b-a)^3}{24 n^2} $$ where $|f(x)''\leq K|$ and $n$ is the number of time steps. if second ...
2
votes
1answer
33 views

If $H$ is positive definite and $s^Ty>0$, then $s^THs-\frac{s^Tyy^Ts}{s^Ty+y^TH^{-1}y}\ne -1$

Let $H\in\mathbb{R}^{n\times n}$ be symmetric and positive definite $s,y\in\mathbb{R}^n$ with $s^Ty>0$ How can we show, that $$s^THs-\frac{s^Tyy^Ts}{s^Ty+y^TH^{-1}y}\ne -1\;?\tag{1}$$ ...
1
vote
5answers
58 views

How to determine if 2 line segments cross?

Give two line segments, each defined by $2$ points in $x,y$ space, such as $L_1 = (x_1,y_1)-(x_2,y_2)$ and $L_2 = (x_3-y_3)-(x_4,y_4)$, and that these points are the result of sampled data (they are ...
0
votes
0answers
29 views

The equivalent of least squares, but for vectors

Given a set of poins, one can use a fitting method such as least squares to find the straight (or the parabola, or the 3rd grade equivalent) that's closest to all points at the same time (via ...
1
vote
0answers
23 views

How can I solve this specific set of equations?

Here are the equations: $$\sum_{k = 1}^n i_k + Y_n u_n = J \quad \quad (1)$$ $$i_1 + Y(u_1 - u_2) = J \quad \quad (2)$$ $$i_k - Y(u_{k - 1} -2u_{k} + u_{k + 1}) = 0, \quad \quad k = 2, ..., n - 2 ...
-4
votes
0answers
33 views

find a method for twin primes and with Golbach conjecture [closed]

There are infinitely many twin primes. Two primes (p, q) are called twin primes if their difference is 2. Let be the number of primes p such that p<= x and p + 2 is also a prime. a sample ...
1
vote
1answer
58 views

Runge-Kutta method for PDE

I consider certain partial differential equation (PDE). For example, let it be heat equation $$u_t = u_{xx}$$ I want to apply numerical Runge-Kutta method for solving it. As a first step I ...
1
vote
0answers
24 views

Numerically stable SVD

In this question regarding SVD, it is explained why eigen decomposition of $ A^tA $ is not numerically stable compared to "direct SVD algorithms". Since the former is the algorithm I'm most familiar ...
1
vote
0answers
33 views

Using Multipule Scale Analysis to solve a non-linear differential equation

I would like to know if there are other methods to solve equations such as this one below. I don't really understand the theory behind the multiple scale analysis and why it works I understand some of ...
-1
votes
1answer
8 views

Computing clock-wise and counter-clock-wise areas of closed X-Y plots [closed]

Given a discrete set of sampled data (a time series of 2 signals) that forms a closed path when plotted against each other (X-Y), and given that the path may cross itself (there may be 2 points on the ...
1
vote
1answer
27 views

Numerically find a potential field from gradient

I know that the gradient of a potential field/scalar field is a vector field, and given the function of the gradient I know how to integrate each component to get back the original scalar field. But ...
0
votes
0answers
16 views

Gauss-Legendre quadrature error

I'm trying to evaluate the error in Gauss-Legendre quadrature formulae on $[a,b]$. So far I have that the error is less or equal to $$ \frac{f^{(2n)}(\xi)}{(2n)!}\langle p_n,p_n \rangle, \enspace ...
0
votes
0answers
11 views

Composite trapezoid rule and trigonometric functions

I am trying to solve the problem talked about in: Trapezoid rule over trigonometric polynomials Show that the composite trapezoid rule over an equidistant partitioning with interval size ...
1
vote
0answers
23 views

How to select the number of nodes in a spline interpolation?

I am writing a program to test the precision of different methods for imputing missing data in a time series. One of the methods I am going to test is a natural cubic spline interpolation. I'll be ...
1
vote
1answer
21 views

Quadrature on segment

Is there a quadrature formula on the segment [0,1], such that on [0,1/2] the points and weights are symetric with respect to 1/4, on [1/2,1] they are symetric with respect to 3/4, and such that the ...
0
votes
0answers
23 views

How many tiles are Symmetrical? [closed]

We have a tape of type $1 * 2015$ had tile from tiles unit square in four different colors so as not exceed two tile of the same color (tile unit square, any tile from type $1*1$) How many tiles are ...
0
votes
0answers
39 views

Is there any practical use of this algorithm?

Example The exact solution of a DE $\frac{dp}{dv}=-1.4\frac{p}{v} $ with initial condition $(P_1,V_1)=(1,1)$ can be obtain by solving the integral ...
0
votes
1answer
14 views

Iteration Scheme Converging

I am in a class called Numerical Analysis and I have a quick question regarding iterative schemes. How would I go about finding out whether or not a certain iteration scheme converges to a unique ...
0
votes
0answers
22 views

Finding area by integration, increasing inaccuracies with complex functions?

I am looking for an explanation as to why the method of integration to find the area of function using limits provides a greater % difference between other methods (In this example Simpsons) with ...
0
votes
0answers
19 views

Error in Gauss-Legendre quadrature

I've tried Googleing this, but so far I haven't succeeded. Can someone point to me a webpage or book in which I can find the error estimate (detailed, not just the final formula) of the Gauss-Legendre ...
2
votes
1answer
20 views

Name and explanation of a Numerical Analysis method for solving systems of non-linear equations

In a non-english textbook of Numerical Analysis there is a method for solving systems of non-linear equations. But not only I can't understand how this method is used but I can't even found the name ...
2
votes
4answers
28 views

Curve fitting of a set of data

Suppose you have a set of data $\{x_i\}$ and $\{y_i\}$ with $i=0,\dots,N$. In order to find two parameters $a,b$ such that the line $$ y=ax+b, $$ give the best linear fit, one proceed minimizing the ...
0
votes
1answer
19 views

Advantages and Limitations of these methods in the subject of Numerical Method [closed]

could you list down the advantages and limitations of these methods? Taylor Series Naive Gaussian Elimination LU Decomposition Dimensional Unconstrained Optimization ( Golden Search Method, ...
2
votes
1answer
37 views

Classifying peak and valley *regions* of a histogram

I've been playing with a few ways of classifying contiguous regions of a histogram as: 1) peak, 2) valley, or 3) in-between bit. Global thresholding has worked minimally well for me so far, but I'm ...
1
vote
1answer
53 views

Finite Difference for Hamilton-Jacobi-Bellman without boundary conditions

Let $t\in\mathbb{R}_+$ denote time, $x \in X$ is the state and $u \in U$ the control. The objective function is $F:X \times U \to\mathbb{R}$ and $f:X \times U \to\mathbb{R}$ is the law of motion for ...
1
vote
1answer
22 views

Show that spectral radius is lower than 1

Consider the following matrix: $I - \frac{1}{h^2}\mu \Delta t A$. Where $A$ is an NxN matrix. The eigenvalues of A $\lambda_j$ are given by $4sin^2(\frac{j\pi}{2(N+1)})$ for $j=1,...,N$. And $\mu , ...
2
votes
1answer
13 views

finite volume methods: what do I have to do with the cell averages after each step?

I'm having a hard time understanding finite volume methods. If I take for example the scalar advection equation $$\partial{u}_{t}+a\partial{u}_{x}=0, a>0$$ with suitable initial and bondary ...
2
votes
0answers
22 views

What is a stochastic differential equation of the form $dZ = f(Z_{prev}, X_{prev})dt + CdW_t$ called?

At every time step I can approximate the change in $Z$ using the following equation: $$ dZ = f(Z_{prev}, X_{prev})dt + CdW_t, \quad(1)$$ $$dW_t = r\sqrt{dt}$$ where $C$ is some constant, and $r$ is ...
0
votes
0answers
8 views

Thresholding in spectra of partial traces of random symmetric matrices

I found an interesting behavior while looking at partial traces of random matrices. This is something I was studying numerically, and I haven't completely ruled out the possibility of numerical ...
2
votes
0answers
15 views

Reference request for finite difference method

I am trying to use finite difference method to solve the minimizing problem $$ J[u]=\min_{u\in BV(Q)}\{\|u-f\|_{L^1(Q)}+|u|_{BV(Q)}\} $$ where $Q=(0,1)\times (0,1)$ is a uint square and ...
2
votes
1answer
21 views

Approximate recursively defined error in fixed point iteration

Problem: With an initial guess of $x_0$, the fixed point iteration is given by $$x_{k+1} = e^{-x_k}, \mbox{ for } k=0,1,2,...$$ If $x^*$ is the exact solution, then the approximation error is $$e_n = ...
0
votes
2answers
17 views

Solving ODE numerically, using derivative steps

Assume I have the ODE $\dot{p}(t) = f(t,p)$, with $p(0) = p_0$ and assume $f(t,p)$ (for simplicity) is only a function of $p$. I want to solve the ODE numerically, using derivative steps, kind of like ...
2
votes
1answer
41 views

How can I solve: $u_{xx} + u_{yy} = g(x,y)$ numerically?

If $u(x,y)$ is defined in $\mathcal{R} = {(x,y): 0 \leq x \leq a, 0 \leq y \leq b}$ $$ u(x,0) = 10 \mbox{ and } u(x,b) = 90, \mbox{ for } 0 \leq x \leq a \\ u(0,y) = 70 \mbox{ and } u(a,y) = ...
1
vote
1answer
88 views

Numerical precision of arctan function

I'm trying to convert points into spherical coordinates, do some filtering/manipulation of the points and convert them back into the Cartesian coordinate frame. These are my transformation equations ...