Multi-Factor Finite Differences

Size: px
Start display at page:

Download "Multi-Factor Finite Differences"

Transcription

1 February 17, 2017

2 Aims and outline Finite differences for more than one direction The θ-method, explicit, implicit, Crank-Nicolson Iterative solution of discretised equations Alternating directions implicit (ADI) methods Analysis of accuracy and stability, Greeks High-dimensional problems, dimension reduction

3 Example: 2-factor basket For simplicity, we will focus on options on (baskets of) traded assets. Consider two stocks ds 1,t = µ 1 S 1,t dt + σ 1 S 1,t dw 1,t, ds 2,t = µ 2 S 2,t dt + σ 2 S 2,t dw 2,t, where the two Wiener processes W 1,t, W 2,t have correlation ρ. An example of a payoff is a put on with expiry T and strike K. S 1,T + S 2,T

4 Black-Scholes PDE A standard hedging argument gives the pricing PDE V t σ2 1S1 2 2 V S1 2 The terminal condition is Boundary conditions are 2 V + ρσ 1 σ 2 S 1 S S 1 S 2 2 σ2 2S2 2 2 V S2 2 V V +rs 1 + rs 2 rv = 0. S 1 S 2 V (S 1, S 2, T ) = max(k S 1 S 2, 0). lim 1, S 2, t) S 1 = 0, lim 1, S 2, t) S 2 = 0.

5 Spatial discretisation Solve on [0, S max ] [0, S max ] [0, T ]. Use time steps t m = m t, m = 0,..., M, t = T /M. Introduce a grid {(j S, k S) : 0 j N, 0 k N}. Denote by V m the approximation to the exact solution V (j S, k S, t m ) at a point (j S, k S), and time t m Vj 1,k+1 m V+1 m V m j+1,k+1 V m j 1,k V m V m j+1,k Vj 1,k 1 m V 1 m V m j+1,k 1

6 Discretisation If we write A for the discrete operator, the backward finite difference equations for the θ-scheme have a similar structure to the 1-D case, V m V m 1 + t [ θa V m 1 ] + (1 θ)a V m = 0. The most obvious choice for the difference approximation is where A V m = A 0V m + A 1V m + A 2V m, A 1 V m = A 2 V m = A 0 V m = 1 [ 2 σ2 1 j2 V m j+1,k 2V m + V m ] 1 [ j 1,k + 2 r j V m j+1,k V m ] m j 1,k rv 1 [ 2 σ2 2 k2 V m +1 2V m + V m ] 1 [ r k V m +1 V m ] 1 1 [ 4 ρσ 1σ 2 jk V m j+1,k+1 V m j 1,k+1 V m j+1,k 1 + V m ] j 1,k 1 or (better) a 7-point-stencil (see later) for A 0.

7 Stencil notation The discretised equations can be written in stencil form as [a] [ V m 1] = [A] [V m ], where [a] = a NW a N a NE a W a C a E a SW a S a SE V j 1,k+1 V +1 V j+1,k+1, [V ] = V j 1,k V V j+1,k V j 1,k 1 V 1 V j+1,k 1 and stencil multiplication produces a scalar by multiplying corresponding entries and summing them up.,

8 Left-hand side a C = 1 + θ t [ σ1 2j 2 + σ2 2k2 + r ] a N = 1 2 θ t [ σ2 2k2 + rk ] a S = 1 2 θ t [ σ2 2k2 rk ] a E = 1 2 θ t [ σ1 2j 2 + rj ] a W = 1 2 θ t [ σ1 2j 2 rj ] a NE = 1 4 θ tρσ 1σ 2 jk a NW = 1 4 θ tρσ 1σ 2 jk a SW = 1 4 θ tρσ 1σ 2 jk a SE = 1 4 θ tρσ 1σ 2 jk

9 Right-hand side A C = 1 (1 θ) t [ σ 2 1 j 2 + σ 2 2 k2 + r ] A N = 1 2 (1 θ) t [ σ 2 2 k2 + rk ] A S = 1 2 (1 θ) t [ σ 2 2 k2 rk ] A E = 1 2 (1 θ) t [ σ 2 1 j 2 + rj ] A W = 1 2 (1 θ) t [ σ 2 1 j 2 rj ] A NE = 1 4 (1 θ) tρσ 1σ 2 jk A NW = 1 4 (1 θ) tρσ 1σ 2 jk A SW = 1 4 (1 θ) tρσ 1σ 2 jk A SE = 1 4 (1 θ) tρσ 1σ 2 jk

10 Intermezzo: 1 factor revisited In the special case of 1 stock, V t σ2 S 2 2 V V + rs rv = 0 S [0, ), t (0, T ). S 2 S Recall the θ-timestepping central difference scheme for the Black-Scholes equation in the form V m 1 V m j j t [ 1 V m 1 j+1 2V m 1 + V m 1 j j 1 S 2 = θ 2 σ2 S 2 j [ 1 + (1 θ) 2 σ2 S 2 Vj+1 m 2V j m j S 2 = θav m 1 + (1 θ)av m j j + V m j 1 m 1 V j+1 + rs j 2 S V m 1 j 1 rv m 1 j m Vj+1 + rs V j 1 m j rv m j 2 S on a grid S 0 = 0, S 1 = S, S 2 = 2 S,..., S N = S max, time steps t 0 = 0, t 1 = t, t 2 = 2 t,..., t M = M t = T, at which we introduce approximations V m n ] ] to V (n S, m t).

11 Practicalities Rewrite as a n V m 1 n 1 + b nv m 1 n + c n V m 1 n+1 = A nv m n 1 + B n V m n + C n V m n+1. In matrix notation, with boundary conditions included, b0 m c m V m 1 a 1 m b1 m c1 m 0 B... 0 V m 1 0 m C m A m 1 B1 m C1 m... 0 = an 1 m bn 1 m cn 1 m V m A m an m bn m N 1 V m 1 N 1 BN 1 m CN 1 m A m N N BN m This is (generally) a tridiagonal linear system, which can be solved directly in O(N) operations in each timestep. V m 0 V m 1... V m N 1 V m N

12 Analysis θ-method The scheme satisfies a discrete maximum principle, if A n, B n, C n 0 A n, B n, C n 1 a n, c n 0 a n + b n + c n 1 ( b n > 0) Setting θ = 0 (explicit scheme), from B n 0 we get stability only for small timesteps. θ = 1 gives the fully implicit method, which is unconditionally stable. Both schemes of first order accurate, Crank-Nicolson second order.

13 von Neumann analysis Since instabilities often appear highly oscillatory, we study the propagation of Fourier modes of the form sin(kx ), cos(kx ). We do this for the constant coefficient equation U t σ2 2 U X 2 + ( r 1 2 σ2) U X ru = 0 on (, ), i. e. X = log S. The θ-scheme for this equation reads ( U m 1 σ 2 ( ) n θ t 2 X 2 U m 1 n+1 2U m 1 n + U m 1 + r σ2 /2 ( ) ) U m 1 n 1 n+1 U m 1 ru m 1 n 1 n = 2 X ( U m σ 2 ( n + (1 θ) t m ) U 2 X 2 n+1 2Um n + r σ 2 ) /2 ( Um m ) n 1 + U n+1 2 X Um m n 1 ru n.

14 von Neumann analysis For U M n = e ikn, first show for m = M that 1 ( U m 2 X n+1 Un 1) m i = X sin k Um n, 1 ( U m X 2 n+1 2Un m + Un 1 m ) 4 = X 2 sin2 (k/2) Un m ; then deduce by induction where U m n = R θ ( X, t; k) M m e ink, R θ ( X, t; k) = 1 (1 θ) t(2σ2 / X 2 sin 2 (k/2) + r i(r 1 2 σ2 )/ X sin k) 1 + θ t(2σ 2 / X 2 sin 2 (k/2) + r i(r 1. 2 σ2 )/ X sin k)

15 von Neumann analysis The scheme is stable, if R θ ( X, t; k) 1 k This is true for all t if θ 0.5 unconditionally stable For θ = 0, need σ 2 t X 2 For θ = 1, For θ = 0.5, lim R 1 = lim R 1 = 0. X 0 t lim X 0 R 1 2 = lim t R 1 2 = 1. High-frequency modes lead to unstable Greeks (use Rannacher start-up).

16 Maximum principles in 2D The 2D scheme satisfies a discrete maximum principle, if A x 0 A x 1 x a x 0, x C a x 1 ( ac > 0) For θ = 1, no constraint on timestep. x For θ < 1, maximum principle only if t = O( S 2 ), but unconditionally stable for θ 1/2. Analysis of drift as for single factor (use upwinding where necessary).

17 7-point stencil Another observation is that for say ρ > 0 the terms a NW a SE, as well as ANW and A SE have the wrong sign. Instead of S for the cross-derivative, we can take the 7-point-stencil 1 2 S 2 [ ] = S 2 [ ] S 2 [ and This is an average of two shifted difference stencils along the direction of the principal component of the diffusion matrix. ].

18 7-point stencil We replace A 0 V m by A 0 V m = 1 [ 2 ρσ 1σ 2 kj V m j+1,k+1 V m +1 V m j+1,k + 2V m V m j 1,k V m 1 + V m ] j 1,k 1 This gives the correct sign for a NE, asw, ANW, ASE. It can be shown that the modified A N, AE etc also have, for constant coefficients and suitable equidistant meshes. For Black-Scholes, can use log-coordinates (or exponentially stretched meshes) to achieve this. The general situation is subject of current research. For ρ [ 1, 0], use points along the diagonal running from NW to SE.

19 von Neumann analysis Consider the model problem u t u 2 x ρ 2 u u x 1 x 2 2 x 2 1 = 0 and its finite difference discretisation. Try solutions of the form u m n,l = R θ ( x, t; k) M m e ink+ilj. If θ 1 2, the scheme is unconditionally stable. If θ < 1 2, stable if (1 2θ)r c, where r = t/ x 2, c depends on the model parameters (here ρ) but not t, x. Similar behaviour for general PDEs.

20 Explicit scheme In the case θ = 0, V m 1 is explicitly computable, = [A] [V m ] starting with the payoff and stepping backwards in time. The stability constraint necessitates a large number of timesteps.

21 Implicit methods The case θ > 0 requires the solution of a potentially large system of equations at each timestep, [a] [ V m 1] = [A] [V m ] = f, the solution of which can be ordered into a single vector [ V m 1 = V0,0 m 1, V1,0 m 1,..., V m 1 N,0, V m 1 0,1, V m 1 1,1,..., V m 1 N,1,. ] V m 1 0,N, V m 1 1,N,..., V m 1 N,N, and the finite difference equations including boundary conditions written as (shorthand V = V m 1 ) KV = f.

22 System matrix The matrix K R (N+1) (N+1) has the shape a0 C a0 E 0 a0 N a NE 0 0 a 1 W a1 C a1 E a1 SN a1 N a NE 1 0 a p S ap SE 0 ap C ap E 0 ap N a NE p K = a p+1 SW ap+1 S ap+1 SE ap+1 W ap+1 C ap+1 E ap+1 NW ap+1 N ap+1 NE 0 a SW p 2 a S 2 p 2 a SE 2 p 2 a W 2 p 2 a C 2 p 2 a E 2 p a SW p 2 a S 1 p 2 a W 1 p 2 a C 1 p 2 1 where al C a C etc, with l = j + kp, p = N + 1, j, k = 0, 1,..., N. This is known as lexicographic ordering.

23 Linear systems Instead of a tridiagonal structure, there are nine non-zero elements per line, not all adjacent ( block-tridiagonal ). N + 1 is the number of points in each direction Gaussian elimination and LU-decomposition lead to fill-in. The complexity increases from O(N) in 1D to O(N 4 ) in 2D! The optimum is O(N 2 ), i.e. proportional to problem size. Need to exploit sparsity pattern, which leads to iterative methods.

24 Relaxation methods To solve: [a] [V ] = f K V = f. Solve by iteration, and let V (n) be the n th iteration for V. Idea: Split K = L + D + U, where L is lower triangular, D diagonal, and U an upper triangular matrix. Define the Jacobi scheme and the Gauss-Seidel scheme D V (n+1) = (L + U)V (n) + f, (D + L)V (n+1) = U V (n) + f.

25 Classical iterations The Jacobi iteration in stencil notation is a NW a N a NE 0 a C [ 0 V (n+1)] = a W 0 a E a SW a S a SE [ V (n)] +f, Gauss-Seidel (GS) with lexicographic ordering a W a C 0 a SW a S a SE [ V (n+1)] = a NW a N a NE 0 0 a E [ V (n)] +f j,

26 Discussion Advantages: exploit sparse matrix structure; easy to implement; Gauss-Seidel proven to converge for positive-definite matrices; can speed up by successive over-relaxation (see lectures) Disadvantages: rate of convergence (good if 1) grows with system size and close to 1 for fine grids; number of iterations required for given accuracy increases rapidly with size. The key word is linear complexity: solution time proportional to number of unknowns. So-called multi-grid solvers can achieve this.

27 American options Projected Gauss-Seidel (PGS) extends as per a C V (n+1/2) a W 0 0 a SW a S a SE [V (n+1)] = a NW a N a NE 0 0 a E [V (n)] + f, followed by V (n+1) ( ) = max V (n+1/2), payoff. Alternatively: use penalty method and faster converging iteration for linear systems (i.e., faster than linear Gauss-Seidel).

28 Splitting schemes The θ-scheme (I θ ta)v m 1 = (1 + (1 θ) ta)v m can be seen as the sequence of an explicit Euler step of size (1 θ) t; an implicit Euler step of size θ t. Idea: Split A = A 0 + A 1 + A 2 into simple components Y 0 = (I + ta 0 )V m (I ta 1 )Y 1 = Y 0 (I ta 2 )Y 2 = Y 1 V m 1 = Y 2 will define a first order accurate, stable solution.

29 Douglas scheme Consider modified discretisations to restore tri-diagonal structure. Idea: Factorise I tθ(a 1 + A 2 + A 0 ) approximately into tri-diagonal factors. Useful to rewrite (I θ ta)v m 1 = (I + (1 θ) ta)v m as (I θ ta)(v m 1 V m ) = ta V m, then any reasonable factorisation is consistent; Approximate I θ t(a 1 + A 2 + A 0 ) = (I tθa 1 )(I θ ta 2 ) θ 2 t 2 A 1 A 2 θ ta 0. Dropping the last two terms, gives the new scheme (I θ ta 1 )(I θ ta 2 )(V m 1 V m ) = ta V m. }{{} =:B

30 Douglas scheme This can alternatively be written as, for general dimension d: Y 0 = V m + ta V m Y j = Y j 1 + θ ta j (Y j V m ) j = 1,..., d V m 1 = Y d The first step is explicit. This is followed by unidirectional implicit corrections. Second order accuracy in t if A 0 = 0 (ρ = 0) and θ = 1 2, first order accurate if ρ 0. For d = 1, the scheme reduces to the well-known θ-scheme.

31 Analysis The main advantage is that the discretisation matrices are tridiagonal (after appropriate re-ordering). What is more, for each j, the system decouples along grid lines. The complexity is therefore O(N 2 ). The order of accuracy is determined by Taylor expansion. Stability analysis by von Neumann. Consider the model problem u t u 2 x ρ 2 u u x 1 x 2 2 x 2 2 = 0.

32 von Neumann stability, d = 2 For un,l M = e ink+ilj : Show, similar to 1D, Bu m = ˆBu m, t Au m = Âu m, where ˆB = (1 + 2rθsk 2 )(1 + 2rθs2 j )  = 2r(sk 2 + s2 j 2ρs k c k s j c j ) with r = t/ x 2, s k = sin(k/2), c k = cos(k/2), etc. Deduce  0 and  2 ˆB if θ 1/2, then for R = ( ˆB Â)/ ˆB, u m 1 = R u m u m. Use Fourier isometry to deduce stable in the l 2 norm.

33 Greeks The Douglas scheme produces stable prices. This is not necessarily true for the sensitivities. The reason is R 1 for t. High frequency errors are not dampened, and blow up when taking finite differences. Can be avoided by Rannacher-type start-up.

34 Rannacher start-up Recall the 1D setting: optimal to replace 2 Crank-Nicolson steps by 4 fully implicit steps of half the step size, Giles and Carter (2006). Idea: Keep A 0 terms explicit, other factors implicit, e.g. (I ta 1 )(I ta 2 )V m 1 = (I + ta 0 )V m. This scheme is strongly stable, and first order accuracy is sufficient for start-up.

35 Craig & Sneyd scheme The scheme corrects for the O( t) error due to the mixed term. Idea: Approximate A 0 V m 1 from earlier using Y d and repeat, (I tθa 1 )(I tθa 2 )(Y d V m ) = tav m (I tθa 1 )(I tθa 2 )(V m 1 V m ) = tav m + θ ta 0 (Y d V m ). This correction step makes the scheme second order accurate in t for θ = 1/2 also for non-zero correlation. Note we have not used any special form of A 0 for the definition of the scheme. The form of A 0 is relevant for stability though.

36 Craig & Sneyd scheme Rewrite as, and generalise to (d 2), Y 0 = V m + tav m Y j = Y j 1 + θ ta j (Y j V m ) j = 1,..., d Ỹ 0 = Y 0 + θ ta 0 (Y d V m ) Ỹ j = Ỹ j 1 + θ ta j (Ỹ j V m ) j = 1,..., d V m 1 = Ỹ d Can be shown to be unconditionally stable in 2D if θ 1 2.

37 Efficiency The complexity of an algorithm is a measure of its computational cost C (in algebraic operations). The ɛ-complexity C(ɛ) is defined as the cost required to achieve a certain accuracy ɛ. In two dimensions, with N mesh points in each direction and M time steps, C = O(MN 2 ). The error of, say Crank-Nicolson and central differences, is O(( S) 2 + ( t) 2 ) or O(M 2 + N 2 ). Therefore, C(ɛ) = O(ɛ 3/2 ). This compares to a Monte Carlo error of M 1 + N 0.5 for N paths and M timesteps, i.e. C(ɛ) = O(ɛ 3 ), or O(ɛ 2 ) if time-stepping can be avoided.

38 High dimensions? In d dimensions, with N mesh points in each direction and M time steps, C = O(MN d ). In terms of the error, C = O(ɛ (d+1)/2 ). This exponential increase of the computational complexity for increasing dimension is referred to as curse of dimensionality. FD asymptotically better than MC up to dimension 3 5. Then use PCA for dimension reduction, exploiting dependency.

39 Literature D. Tavella, C. Randall: Pricing Financial Instruments The Finite Difference Method. Wiley, M. B. Giles, R. Carter: Convergence analysis of Crank-Nicolson and Rannacher time-marching. Journal of Computational Finance, 9(4), I. J. D. Craig, A. D. Sneyd: An alternating-direction implicit scheme for parabolic equations with mixed derivatives, Comput. Math. Applic., 16(4), Karel in t Hout: ADI schemes in the Numerical Solution of the Heston PDE, Proceedings of the International Conference Numerical Analysis and Applied Mathematics, 2007.

The method of lines (MOL) for the diffusion equation

The method of lines (MOL) for the diffusion equation Chapter 1 The method of lines (MOL) for the diffusion equation The method of lines refers to an approximation of one or more partial differential equations with ordinary differential equations in just

More information

ME Computational Fluid Mechanics Lecture 5

ME Computational Fluid Mechanics Lecture 5 ME - 733 Computational Fluid Mechanics Lecture 5 Dr./ Ahmed Nagib Elmekawy Dec. 20, 2018 Elliptic PDEs: Finite Difference Formulation Using central difference formulation, the so called five-point formula

More information

Radial Basis Functions generated Finite Differences (RBF-FD) for Solving High-Dimensional PDEs in Finance

Radial Basis Functions generated Finite Differences (RBF-FD) for Solving High-Dimensional PDEs in Finance 1 / 27 Radial Basis Functions generated Finite Differences (RBF-FD) for Solving High-Dimensional PDEs in Finance S. Milovanović, L. von Sydow Uppsala University Department of Information Technology Division

More information

Today s class. Linear Algebraic Equations LU Decomposition. Numerical Methods, Fall 2011 Lecture 8. Prof. Jinbo Bi CSE, UConn

Today s class. Linear Algebraic Equations LU Decomposition. Numerical Methods, Fall 2011 Lecture 8. Prof. Jinbo Bi CSE, UConn Today s class Linear Algebraic Equations LU Decomposition 1 Linear Algebraic Equations Gaussian Elimination works well for solving linear systems of the form: AX = B What if you have to solve the linear

More information

Numerical Solution Techniques in Mechanical and Aerospace Engineering

Numerical Solution Techniques in Mechanical and Aerospace Engineering Numerical Solution Techniques in Mechanical and Aerospace Engineering Chunlei Liang LECTURE 3 Solvers of linear algebraic equations 3.1. Outline of Lecture Finite-difference method for a 2D elliptic PDE

More information

Math 471 (Numerical methods) Chapter 3 (second half). System of equations

Math 471 (Numerical methods) Chapter 3 (second half). System of equations Math 47 (Numerical methods) Chapter 3 (second half). System of equations Overlap 3.5 3.8 of Bradie 3.5 LU factorization w/o pivoting. Motivation: ( ) A I Gaussian Elimination (U L ) where U is upper triangular

More information

FDM for parabolic equations

FDM for parabolic equations FDM for parabolic equations Consider the heat equation where Well-posed problem Existence & Uniqueness Mass & Energy decreasing FDM for parabolic equations CNFD Crank-Nicolson + 2 nd order finite difference

More information

6. Iterative Methods for Linear Systems. The stepwise approach to the solution...

6. Iterative Methods for Linear Systems. The stepwise approach to the solution... 6 Iterative Methods for Linear Systems The stepwise approach to the solution Miriam Mehl: 6 Iterative Methods for Linear Systems The stepwise approach to the solution, January 18, 2013 1 61 Large Sparse

More information

Iterative Methods. Splitting Methods

Iterative Methods. Splitting Methods Iterative Methods Splitting Methods 1 Direct Methods Solving Ax = b using direct methods. Gaussian elimination (using LU decomposition) Variants of LU, including Crout and Doolittle Other decomposition

More information

Introduction to numerical schemes

Introduction to numerical schemes 236861 Numerical Geometry of Images Tutorial 2 Introduction to numerical schemes Heat equation The simple parabolic PDE with the initial values u t = K 2 u 2 x u(0, x) = u 0 (x) and some boundary conditions

More information

JACOBI S ITERATION METHOD

JACOBI S ITERATION METHOD ITERATION METHODS These are methods which compute a sequence of progressively accurate iterates to approximate the solution of Ax = b. We need such methods for solving many large linear systems. Sometimes

More information

Introduction to PDEs and Numerical Methods: Exam 1

Introduction to PDEs and Numerical Methods: Exam 1 Prof Dr Thomas Sonar, Institute of Analysis Winter Semester 2003/4 17122003 Introduction to PDEs and Numerical Methods: Exam 1 To obtain full points explain your solutions thoroughly and self-consistently

More information

Chapter Two: Numerical Methods for Elliptic PDEs. 1 Finite Difference Methods for Elliptic PDEs

Chapter Two: Numerical Methods for Elliptic PDEs. 1 Finite Difference Methods for Elliptic PDEs Chapter Two: Numerical Methods for Elliptic PDEs Finite Difference Methods for Elliptic PDEs.. Finite difference scheme. We consider a simple example u := subject to Dirichlet boundary conditions ( ) u

More information

9. Iterative Methods for Large Linear Systems

9. Iterative Methods for Large Linear Systems EE507 - Computational Techniques for EE Jitkomut Songsiri 9. Iterative Methods for Large Linear Systems introduction splitting method Jacobi method Gauss-Seidel method successive overrelaxation (SOR) 9-1

More information

Lecture 4.5 Schemes for Parabolic Type Equations

Lecture 4.5 Schemes for Parabolic Type Equations Lecture 4.5 Schemes for Parabolic Type Equations 1 Difference Schemes for Parabolic Equations One-dimensional problems: Consider the unsteady diffusion problem (parabolic in nature) in a thin wire governed

More information

Partial Differential Equations

Partial Differential Equations Partial Differential Equations Introduction Deng Li Discretization Methods Chunfang Chen, Danny Thorne, Adam Zornes CS521 Feb.,7, 2006 What do You Stand For? A PDE is a Partial Differential Equation This

More information

High-order ADI schemes for convection-diffusion equations with mixed derivative terms

High-order ADI schemes for convection-diffusion equations with mixed derivative terms High-order ADI schemes for convection-diffusion equations with mixed derivative terms B. Düring, M. Fournié and A. Rigal Abstract We consider new high-order Alternating Direction Implicit ADI) schemes

More information

Introduction to PDEs and Numerical Methods Lecture 7. Solving linear systems

Introduction to PDEs and Numerical Methods Lecture 7. Solving linear systems Platzhalter für Bild, Bild auf Titelfolie hinter das Logo einsetzen Introduction to PDEs and Numerical Methods Lecture 7. Solving linear systems Dr. Noemi Friedman, 09.2.205. Reminder: Instationary heat

More information

Numerical Solutions to Partial Differential Equations

Numerical Solutions to Partial Differential Equations Numerical Solutions to Partial Differential Equations Zhiping Li LMAM and School of Mathematical Sciences Peking University A Model Problem and Its Difference Approximations 1-D Initial Boundary Value

More information

Finite difference methods for the diffusion equation

Finite difference methods for the diffusion equation Finite difference methods for the diffusion equation D150, Tillämpade numeriska metoder II Olof Runborg May 0, 003 These notes summarize a part of the material in Chapter 13 of Iserles. They are based

More information

Department of Mathematics California State University, Los Angeles Master s Degree Comprehensive Examination in. NUMERICAL ANALYSIS Spring 2015

Department of Mathematics California State University, Los Angeles Master s Degree Comprehensive Examination in. NUMERICAL ANALYSIS Spring 2015 Department of Mathematics California State University, Los Angeles Master s Degree Comprehensive Examination in NUMERICAL ANALYSIS Spring 2015 Instructions: Do exactly two problems from Part A AND two

More information

Introduction to PDEs and Numerical Methods Tutorial 5. Finite difference methods equilibrium equation and iterative solvers

Introduction to PDEs and Numerical Methods Tutorial 5. Finite difference methods equilibrium equation and iterative solvers Platzhalter für Bild, Bild auf Titelfolie hinter das Logo einsetzen Introduction to PDEs and Numerical Methods Tutorial 5. Finite difference methods equilibrium equation and iterative solvers Dr. Noemi

More information

Finite Difference Method for PDE. Y V S S Sanyasiraju Professor, Department of Mathematics IIT Madras, Chennai 36

Finite Difference Method for PDE. Y V S S Sanyasiraju Professor, Department of Mathematics IIT Madras, Chennai 36 Finite Difference Method for PDE Y V S S Sanyasiraju Professor, Department of Mathematics IIT Madras, Chennai 36 1 Classification of the Partial Differential Equations Consider a scalar second order partial

More information

Tutorial 2. Introduction to numerical schemes

Tutorial 2. Introduction to numerical schemes 236861 Numerical Geometry of Images Tutorial 2 Introduction to numerical schemes c 2012 Classifying PDEs Looking at the PDE Au xx + 2Bu xy + Cu yy + Du x + Eu y + Fu +.. = 0, and its discriminant, B 2

More information

Algebra C Numerical Linear Algebra Sample Exam Problems

Algebra C Numerical Linear Algebra Sample Exam Problems Algebra C Numerical Linear Algebra Sample Exam Problems Notation. Denote by V a finite-dimensional Hilbert space with inner product (, ) and corresponding norm. The abbreviation SPD is used for symmetric

More information

Review of matrices. Let m, n IN. A rectangle of numbers written like A =

Review of matrices. Let m, n IN. A rectangle of numbers written like A = Review of matrices Let m, n IN. A rectangle of numbers written like a 11 a 12... a 1n a 21 a 22... a 2n A =...... a m1 a m2... a mn where each a ij IR is called a matrix with m rows and n columns or an

More information

Multigrid Methods and their application in CFD

Multigrid Methods and their application in CFD Multigrid Methods and their application in CFD Michael Wurst TU München 16.06.2009 1 Multigrid Methods Definition Multigrid (MG) methods in numerical analysis are a group of algorithms for solving differential

More information

Kasetsart University Workshop. Multigrid methods: An introduction

Kasetsart University Workshop. Multigrid methods: An introduction Kasetsart University Workshop Multigrid methods: An introduction Dr. Anand Pardhanani Mathematics Department Earlham College Richmond, Indiana USA pardhan@earlham.edu A copy of these slides is available

More information

STABILITY FOR PARABOLIC SOLVERS

STABILITY FOR PARABOLIC SOLVERS Review STABILITY FOR PARABOLIC SOLVERS School of Mathematics Semester 1 2008 OUTLINE Review 1 REVIEW 2 STABILITY: EXPLICIT METHOD Explicit Method as a Matrix Equation Growing Errors Stability Constraint

More information

NUMERICAL METHODS FOR ENGINEERING APPLICATION

NUMERICAL METHODS FOR ENGINEERING APPLICATION NUMERICAL METHODS FOR ENGINEERING APPLICATION Second Edition JOEL H. FERZIGER A Wiley-Interscience Publication JOHN WILEY & SONS, INC. New York / Chichester / Weinheim / Brisbane / Singapore / Toronto

More information

Iterative Methods for Linear Systems

Iterative Methods for Linear Systems Iterative Methods for Linear Systems 1. Introduction: Direct solvers versus iterative solvers In many applications we have to solve a linear system Ax = b with A R n n and b R n given. If n is large the

More information

Solving linear systems (6 lectures)

Solving linear systems (6 lectures) Chapter 2 Solving linear systems (6 lectures) 2.1 Solving linear systems: LU factorization (1 lectures) Reference: [Trefethen, Bau III] Lecture 20, 21 How do you solve Ax = b? (2.1.1) In numerical linear

More information

2.29 Numerical Fluid Mechanics Spring 2015 Lecture 13

2.29 Numerical Fluid Mechanics Spring 2015 Lecture 13 REVIEW Lecture 12: Spring 2015 Lecture 13 Grid-Refinement and Error estimation Estimation of the order of convergence and of the discretization error Richardson s extrapolation and Iterative improvements

More information

ECE539 - Advanced Theory of Semiconductors and Semiconductor Devices. Numerical Methods and Simulation / Umberto Ravaioli

ECE539 - Advanced Theory of Semiconductors and Semiconductor Devices. Numerical Methods and Simulation / Umberto Ravaioli ECE539 - Advanced Theory of Semiconductors and Semiconductor Devices 1 General concepts Numerical Methods and Simulation / Umberto Ravaioli Introduction to the Numerical Solution of Partial Differential

More information

Lecture 16 Methods for System of Linear Equations (Linear Systems) Songting Luo. Department of Mathematics Iowa State University

Lecture 16 Methods for System of Linear Equations (Linear Systems) Songting Luo. Department of Mathematics Iowa State University Lecture 16 Methods for System of Linear Equations (Linear Systems) Songting Luo Department of Mathematics Iowa State University MATH 481 Numerical Methods for Differential Equations Songting Luo ( Department

More information

Computation Fluid Dynamics

Computation Fluid Dynamics Computation Fluid Dynamics CFD I Jitesh Gajjar Maths Dept Manchester University Computation Fluid Dynamics p.1/189 Garbage In, Garbage Out We will begin with a discussion of errors. Useful to understand

More information

Finite Difference Methods (FDMs) 2

Finite Difference Methods (FDMs) 2 Finite Difference Methods (FDMs) 2 Time- dependent PDEs A partial differential equation of the form (15.1) where A, B, and C are constants, is called quasilinear. There are three types of quasilinear equations:

More information

Numerical Solution of partial differential equations

Numerical Solution of partial differential equations G. D. SMITH Brunei University Numerical Solution of partial differential equations FINITE DIFFERENCE METHODS THIRD EDITION CLARENDON PRESS OXFORD Contents NOTATION 1. INTRODUCTION AND FINITE-DIFFERENCE

More information

Next topics: Solving systems of linear equations

Next topics: Solving systems of linear equations Next topics: Solving systems of linear equations 1 Gaussian elimination (today) 2 Gaussian elimination with partial pivoting (Week 9) 3 The method of LU-decomposition (Week 10) 4 Iterative techniques:

More information

Lecture 9 Approximations of Laplace s Equation, Finite Element Method. Mathématiques appliquées (MATH0504-1) B. Dewals, C.

Lecture 9 Approximations of Laplace s Equation, Finite Element Method. Mathématiques appliquées (MATH0504-1) B. Dewals, C. Lecture 9 Approximations of Laplace s Equation, Finite Element Method Mathématiques appliquées (MATH54-1) B. Dewals, C. Geuzaine V1.2 23/11/218 1 Learning objectives of this lecture Apply the finite difference

More information

Aspects of Multigrid

Aspects of Multigrid Aspects of Multigrid Kees Oosterlee 1,2 1 Delft University of Technology, Delft. 2 CWI, Center for Mathematics and Computer Science, Amsterdam, SIAM Chapter Workshop Day, May 30th 2018 C.W.Oosterlee (CWI)

More information

Applied Mathematics 205. Unit III: Numerical Calculus. Lecturer: Dr. David Knezevic

Applied Mathematics 205. Unit III: Numerical Calculus. Lecturer: Dr. David Knezevic Applied Mathematics 205 Unit III: Numerical Calculus Lecturer: Dr. David Knezevic Unit III: Numerical Calculus Chapter III.3: Boundary Value Problems and PDEs 2 / 96 ODE Boundary Value Problems 3 / 96

More information

AIMS Exercise Set # 1

AIMS Exercise Set # 1 AIMS Exercise Set #. Determine the form of the single precision floating point arithmetic used in the computers at AIMS. What is the largest number that can be accurately represented? What is the smallest

More information

Index. higher order methods, 52 nonlinear, 36 with variable coefficients, 34 Burgers equation, 234 BVP, see boundary value problems

Index. higher order methods, 52 nonlinear, 36 with variable coefficients, 34 Burgers equation, 234 BVP, see boundary value problems Index A-conjugate directions, 83 A-stability, 171 A( )-stability, 171 absolute error, 243 absolute stability, 149 for systems of equations, 154 absorbing boundary conditions, 228 Adams Bashforth methods,

More information

Numerical Solutions to Partial Differential Equations

Numerical Solutions to Partial Differential Equations Numerical Solutions to Partial Differential Equations Zhiping Li LMAM and School of Mathematical Sciences Peking University A Model Problem in a 2D Box Region Let us consider a model problem of parabolic

More information

No USING LOCALISED QUADRATIC FUNCTIONS ON AN IRREGULAR GRID FOR PRICING HIGH-DIMENSIONAL AMERICAN OPTIONS. By S.J. Berridge, J.M.

No USING LOCALISED QUADRATIC FUNCTIONS ON AN IRREGULAR GRID FOR PRICING HIGH-DIMENSIONAL AMERICAN OPTIONS. By S.J. Berridge, J.M. No. 4 USING LOCALISED QUADRATIC FUNCTIONS ON AN IRREGULAR GRID FOR PRICING HIGH-DIMENSIONAL AMERICAN OPTIONS By S.J. Berridge, J.M. Schumacher February 4 ISSN 94-785 Using localised quadratic functions

More information

Block-Structured Adaptive Mesh Refinement

Block-Structured Adaptive Mesh Refinement Block-Structured Adaptive Mesh Refinement Lecture 2 Incompressible Navier-Stokes Equations Fractional Step Scheme 1-D AMR for classical PDE s hyperbolic elliptic parabolic Accuracy considerations Bell

More information

Boundary Value Problems and Iterative Methods for Linear Systems

Boundary Value Problems and Iterative Methods for Linear Systems Boundary Value Problems and Iterative Methods for Linear Systems 1. Equilibrium Problems 1.1. Abstract setting We want to find a displacement u V. Here V is a complete vector space with a norm v V. In

More information

Lecture 1. Finite difference and finite element methods. Partial differential equations (PDEs) Solving the heat equation numerically

Lecture 1. Finite difference and finite element methods. Partial differential equations (PDEs) Solving the heat equation numerically Finite difference and finite element methods Lecture 1 Scope of the course Analysis and implementation of numerical methods for pricing options. Models: Black-Scholes, stochastic volatility, exponential

More information

Lecture Notes on Numerical Schemes for Flow and Transport Problems

Lecture Notes on Numerical Schemes for Flow and Transport Problems Lecture Notes on Numerical Schemes for Flow and Transport Problems by Sri Redeki Pudaprasetya sr pudap@math.itb.ac.id Department of Mathematics Faculty of Mathematics and Natural Sciences Bandung Institute

More information

Review for Exam 2 Ben Wang and Mark Styczynski

Review for Exam 2 Ben Wang and Mark Styczynski Review for Exam Ben Wang and Mark Styczynski This is a rough approximation of what we went over in the review session. This is actually more detailed in portions than what we went over. Also, please note

More information

Numerical Analysis of Differential Equations Numerical Solution of Parabolic Equations

Numerical Analysis of Differential Equations Numerical Solution of Parabolic Equations Numerical Analysis of Differential Equations 215 6 Numerical Solution of Parabolic Equations 6 Numerical Solution of Parabolic Equations TU Bergakademie Freiberg, SS 2012 Numerical Analysis of Differential

More information

Numerical Methods I Non-Square and Sparse Linear Systems

Numerical Methods I Non-Square and Sparse Linear Systems Numerical Methods I Non-Square and Sparse Linear Systems Aleksandar Donev Courant Institute, NYU 1 donev@courant.nyu.edu 1 MATH-GA 2011.003 / CSCI-GA 2945.003, Fall 2014 September 25th, 2014 A. Donev (Courant

More information

Finite Differences for Differential Equations 28 PART II. Finite Difference Methods for Differential Equations

Finite Differences for Differential Equations 28 PART II. Finite Difference Methods for Differential Equations Finite Differences for Differential Equations 28 PART II Finite Difference Methods for Differential Equations Finite Differences for Differential Equations 29 BOUNDARY VALUE PROBLEMS (I) Solving a TWO

More information

The Solution of Linear Systems AX = B

The Solution of Linear Systems AX = B Chapter 2 The Solution of Linear Systems AX = B 21 Upper-triangular Linear Systems We will now develop the back-substitution algorithm, which is useful for solving a linear system of equations that has

More information

12 The Heat equation in one spatial dimension: Simple explicit method and Stability analysis

12 The Heat equation in one spatial dimension: Simple explicit method and Stability analysis ATH 337, by T. Lakoba, University of Vermont 113 12 The Heat equation in one spatial dimension: Simple explicit method and Stability analysis 12.1 Formulation of the IBVP and the minimax property of its

More information

Lecture Notes on Numerical Schemes for Flow and Transport Problems

Lecture Notes on Numerical Schemes for Flow and Transport Problems Lecture Notes on Numerical Schemes for Flow and Transport Problems by Sri Redeki Pudaprasetya sr pudap@math.itb.ac.id Department of Mathematics Faculty of Mathematics and Natural Sciences Bandung Institute

More information

Chapter 2. Solving Systems of Equations. 2.1 Gaussian elimination

Chapter 2. Solving Systems of Equations. 2.1 Gaussian elimination Chapter 2 Solving Systems of Equations A large number of real life applications which are resolved through mathematical modeling will end up taking the form of the following very simple looking matrix

More information

Scientific Computing

Scientific Computing Scientific Computing Direct solution methods Martin van Gijzen Delft University of Technology October 3, 2018 1 Program October 3 Matrix norms LU decomposition Basic algorithm Cost Stability Pivoting Pivoting

More information

Block-tridiagonal matrices

Block-tridiagonal matrices Block-tridiagonal matrices. p.1/31 Block-tridiagonal matrices - where do these arise? - as a result of a particular mesh-point ordering - as a part of a factorization procedure, for example when we compute

More information

Exponential integration of large systems of ODEs

Exponential integration of large systems of ODEs Exponential integration of large systems of ODEs Jitse Niesen (University of Leeds) in collaboration with Will Wright (Melbourne University) 23rd Biennial Conference on Numerical Analysis, June 2009 Plan

More information

LU Factorization. Marco Chiarandini. DM559 Linear and Integer Programming. Department of Mathematics & Computer Science University of Southern Denmark

LU Factorization. Marco Chiarandini. DM559 Linear and Integer Programming. Department of Mathematics & Computer Science University of Southern Denmark DM559 Linear and Integer Programming LU Factorization Marco Chiarandini Department of Mathematics & Computer Science University of Southern Denmark [Based on slides by Lieven Vandenberghe, UCLA] Outline

More information

arxiv: v1 [math.na] 5 May 2012

arxiv: v1 [math.na] 5 May 2012 arxiv:205.63v [math.na] 5 May 202 Stability of ADI schemes for multidimensional diffusion equations with mixed derivative terms K. J. in t Hout and C. Mishra April 30, 202 Abstract In this paper the unconditional

More information

A fast and well-conditioned spectral method: The US method

A fast and well-conditioned spectral method: The US method A fast and well-conditioned spectral method: The US method Alex Townsend University of Oxford Leslie Fox Prize, 24th of June 23 Partially based on: S. Olver & T., A fast and well-conditioned spectral method,

More information

Lecture 9: Numerical Linear Algebra Primer (February 11st)

Lecture 9: Numerical Linear Algebra Primer (February 11st) 10-725/36-725: Convex Optimization Spring 2015 Lecture 9: Numerical Linear Algebra Primer (February 11st) Lecturer: Ryan Tibshirani Scribes: Avinash Siravuru, Guofan Wu, Maosheng Liu Note: LaTeX template

More information

Lecture 2: Computing functions of dense matrices

Lecture 2: Computing functions of dense matrices Lecture 2: Computing functions of dense matrices Paola Boito and Federico Poloni Università di Pisa Pisa - Hokkaido - Roma2 Summer School Pisa, August 27 - September 8, 2018 Introduction In this lecture

More information

Boundary Value Problems - Solving 3-D Finite-Difference problems Jacob White

Boundary Value Problems - Solving 3-D Finite-Difference problems Jacob White Introduction to Simulation - Lecture 2 Boundary Value Problems - Solving 3-D Finite-Difference problems Jacob White Thanks to Deepak Ramaswamy, Michal Rewienski, and Karen Veroy Outline Reminder about

More information

Numerical Algorithms for Visual Computing II 2010/11 Example Solutions for Assignment 6

Numerical Algorithms for Visual Computing II 2010/11 Example Solutions for Assignment 6 Numerical Algorithms for Visual Computing II 00/ Example Solutions for Assignment 6 Problem (Matrix Stability Infusion). The matrix A of the arising matrix notation U n+ = AU n takes the following form,

More information

CLASSICAL ITERATIVE METHODS

CLASSICAL ITERATIVE METHODS CLASSICAL ITERATIVE METHODS LONG CHEN In this notes we discuss classic iterative methods on solving the linear operator equation (1) Au = f, posed on a finite dimensional Hilbert space V = R N equipped

More information

Krylov methods for the computation of matrix functions

Krylov methods for the computation of matrix functions Krylov methods for the computation of matrix functions Jitse Niesen (University of Leeds) in collaboration with Will Wright (Melbourne University) Heriot-Watt University, March 2010 Outline Definition

More information

q t = F q x. (1) is a flux of q due to diffusion. Although very complex parameterizations for F q

q t = F q x. (1) is a flux of q due to diffusion. Although very complex parameterizations for F q ! Revised Tuesday, December 8, 015! 1 Chapter 7: Diffusion Copyright 015, David A. Randall 7.1! Introduction Diffusion is a macroscopic statistical description of microscopic advection. Here microscopic

More information

AMS526: Numerical Analysis I (Numerical Linear Algebra for Computational and Data Sciences)

AMS526: Numerical Analysis I (Numerical Linear Algebra for Computational and Data Sciences) AMS526: Numerical Analysis I (Numerical Linear Algebra for Computational and Data Sciences) Lecture 19: Computing the SVD; Sparse Linear Systems Xiangmin Jiao Stony Brook University Xiangmin Jiao Numerical

More information

Using pseudo-parabolic and fractional equations for option pricing in jump diffusion models

Using pseudo-parabolic and fractional equations for option pricing in jump diffusion models Using pseudo-parabolic and fractional equations for option pricing in jump diffusion models Andrey Itkin 1 1 HAP Capital LLC and NYU Poly joined work with Peter Carr) 6-th World Bachelier Congress, Toronto,

More information

Computer Aided Design of Thermal Systems (ME648)

Computer Aided Design of Thermal Systems (ME648) Computer Aided Design of Thermal Systems (ME648) PG/Open Elective Credits: 3-0-0-9 Updated Syallabus: Introduction. Basic Considerations in Design. Modelling of Thermal Systems. Numerical Modelling and

More information

Sparse Linear Systems. Iterative Methods for Sparse Linear Systems. Motivation for Studying Sparse Linear Systems. Partial Differential Equations

Sparse Linear Systems. Iterative Methods for Sparse Linear Systems. Motivation for Studying Sparse Linear Systems. Partial Differential Equations Sparse Linear Systems Iterative Methods for Sparse Linear Systems Matrix Computations and Applications, Lecture C11 Fredrik Bengzon, Robert Söderlund We consider the problem of solving the linear system

More information

An Efficient Algorithm Based on Quadratic Spline Collocation and Finite Difference Methods for Parabolic Partial Differential Equations.

An Efficient Algorithm Based on Quadratic Spline Collocation and Finite Difference Methods for Parabolic Partial Differential Equations. An Efficient Algorithm Based on Quadratic Spline Collocation and Finite Difference Methods for Parabolic Partial Differential Equations by Tong Chen A thesis submitted in conformity with the requirements

More information

Platzhalter für Bild, Bild auf Titelfolie hinter das Logo einsetzen

Platzhalter für Bild, Bild auf Titelfolie hinter das Logo einsetzen Platzhalter für Bild, Bild auf Titelfolie hinter das Logo einsetzen Introduction to PDEs and Numerical Methods Lecture 6: Numerical solution of the heat equation with FD method: method of lines, Euler

More information

Numerical Linear Algebra Primer. Ryan Tibshirani Convex Optimization /36-725

Numerical Linear Algebra Primer. Ryan Tibshirani Convex Optimization /36-725 Numerical Linear Algebra Primer Ryan Tibshirani Convex Optimization 10-725/36-725 Last time: proximal gradient descent Consider the problem min g(x) + h(x) with g, h convex, g differentiable, and h simple

More information

Dynamic Risk Measures and Nonlinear Expectations with Markov Chain noise

Dynamic Risk Measures and Nonlinear Expectations with Markov Chain noise Dynamic Risk Measures and Nonlinear Expectations with Markov Chain noise Robert J. Elliott 1 Samuel N. Cohen 2 1 Department of Commerce, University of South Australia 2 Mathematical Insitute, University

More information

Finite difference method for heat equation

Finite difference method for heat equation Finite difference method for heat equation Praveen. C praveen@math.tifrbng.res.in Tata Institute of Fundamental Research Center for Applicable Mathematics Bangalore 560065 http://math.tifrbng.res.in/~praveen

More information

Classical iterative methods for linear systems

Classical iterative methods for linear systems Classical iterative methods for linear systems Ed Bueler MATH 615 Numerical Analysis of Differential Equations 27 February 1 March, 2017 Ed Bueler (MATH 615 NADEs) Classical iterative methods for linear

More information

Characteristic finite-difference solution Stability of C C (CDS in time/space, explicit): Example: Effective numerical wave numbers and dispersion

Characteristic finite-difference solution Stability of C C (CDS in time/space, explicit): Example: Effective numerical wave numbers and dispersion Spring 015 Lecture 14 REVIEW Lecture 13: Stability: Von Neumann Ex.: 1st order linear convection/wave eqn., F-B scheme Hyperbolic PDEs and Stability nd order wave equation and waves on a string Characteristic

More information

Linear Solvers. Andrew Hazel

Linear Solvers. Andrew Hazel Linear Solvers Andrew Hazel Introduction Thus far we have talked about the formulation and discretisation of physical problems...... and stopped when we got to a discrete linear system of equations. Introduction

More information

Course Notes: Week 1

Course Notes: Week 1 Course Notes: Week 1 Math 270C: Applied Numerical Linear Algebra 1 Lecture 1: Introduction (3/28/11) We will focus on iterative methods for solving linear systems of equations (and some discussion of eigenvalues

More information

AMS526: Numerical Analysis I (Numerical Linear Algebra)

AMS526: Numerical Analysis I (Numerical Linear Algebra) AMS526: Numerical Analysis I (Numerical Linear Algebra) Lecture 24: Preconditioning and Multigrid Solver Xiangmin Jiao SUNY Stony Brook Xiangmin Jiao Numerical Analysis I 1 / 5 Preconditioning Motivation:

More information

Stabilization and Acceleration of Algebraic Multigrid Method

Stabilization and Acceleration of Algebraic Multigrid Method Stabilization and Acceleration of Algebraic Multigrid Method Recursive Projection Algorithm A. Jemcov J.P. Maruszewski Fluent Inc. October 24, 2006 Outline 1 Need for Algorithm Stabilization and Acceleration

More information

u n 2 4 u n 36 u n 1, n 1.

u n 2 4 u n 36 u n 1, n 1. Exercise 1 Let (u n ) be the sequence defined by Set v n = u n 1 x+ u n and f (x) = 4 x. 1. Solve the equations f (x) = 1 and f (x) =. u 0 = 0, n Z +, u n+1 = u n + 4 u n.. Prove that if u n < 1, then

More information

1.6: 16, 20, 24, 27, 28

1.6: 16, 20, 24, 27, 28 .6: 6, 2, 24, 27, 28 6) If A is positive definite, then A is positive definite. The proof of the above statement can easily be shown for the following 2 2 matrix, a b A = b c If that matrix is positive

More information

Iterative Methods and Multigrid

Iterative Methods and Multigrid Iterative Methods and Multigrid Part 1: Introduction to Multigrid 2000 Eric de Sturler 1 12/02/09 MG01.prz Basic Iterative Methods (1) Nonlinear equation: f(x) = 0 Rewrite as x = F(x), and iterate x i+1

More information

Computational Linear Algebra

Computational Linear Algebra Computational Linear Algebra PD Dr. rer. nat. habil. Ralf-Peter Mundani Computation in Engineering / BGU Scientific Computing in Computer Science / INF Winter Term 2018/19 Part 4: Iterative Methods PD

More information

Finite Difference Methods for

Finite Difference Methods for CE 601: Numerical Methods Lecture 33 Finite Difference Methods for PDEs Course Coordinator: Course Coordinator: Dr. Suresh A. Kartha, Associate Professor, Department of Civil Engineering, IIT Guwahati.

More information

Diffusion / Parabolic Equations. PHY 688: Numerical Methods for (Astro)Physics

Diffusion / Parabolic Equations. PHY 688: Numerical Methods for (Astro)Physics Diffusion / Parabolic Equations Summary of PDEs (so far...) Hyperbolic Think: advection Real, finite speed(s) at which information propagates carries changes in the solution Second-order explicit methods

More information

Lab 1: Iterative Methods for Solving Linear Systems

Lab 1: Iterative Methods for Solving Linear Systems Lab 1: Iterative Methods for Solving Linear Systems January 22, 2017 Introduction Many real world applications require the solution to very large and sparse linear systems where direct methods such as

More information

Practical Linear Algebra: A Geometry Toolbox

Practical Linear Algebra: A Geometry Toolbox Practical Linear Algebra: A Geometry Toolbox Third edition Chapter 12: Gauss for Linear Systems Gerald Farin & Dianne Hansford CRC Press, Taylor & Francis Group, An A K Peters Book www.farinhansford.com/books/pla

More information

Stability of the Parareal Algorithm

Stability of the Parareal Algorithm Stability of the Parareal Algorithm Gunnar Andreas Staff and Einar M. Rønquist Norwegian University of Science and Technology Department of Mathematical Sciences Summary. We discuss the stability of the

More information

Elliptic Problems / Multigrid. PHY 604: Computational Methods for Physics and Astrophysics II

Elliptic Problems / Multigrid. PHY 604: Computational Methods for Physics and Astrophysics II Elliptic Problems / Multigrid Summary of Hyperbolic PDEs We looked at a simple linear and a nonlinear scalar hyperbolic PDE There is a speed associated with the change of the solution Explicit methods

More information

Numerical accuracy in the solution of the shallow-water equations

Numerical accuracy in the solution of the shallow-water equations Numerical accuracy in the solution of the shallow-water equations P. Broomans & C. Vuik Delft University of Technology, Delft, Netherlands A.E. Mynett & J. Mooiman WL Delft Hydraulics, Delft, Netherlands

More information

7 Hyperbolic Differential Equations

7 Hyperbolic Differential Equations Numerical Analysis of Differential Equations 243 7 Hyperbolic Differential Equations While parabolic equations model diffusion processes, hyperbolic equations model wave propagation and transport phenomena.

More information

Numerical Linear Algebra

Numerical Linear Algebra Numerical Linear Algebra Decompositions, numerical aspects Gerard Sleijpen and Martin van Gijzen September 27, 2017 1 Delft University of Technology Program Lecture 2 LU-decomposition Basic algorithm Cost

More information

Program Lecture 2. Numerical Linear Algebra. Gaussian elimination (2) Gaussian elimination. Decompositions, numerical aspects

Program Lecture 2. Numerical Linear Algebra. Gaussian elimination (2) Gaussian elimination. Decompositions, numerical aspects Numerical Linear Algebra Decompositions, numerical aspects Program Lecture 2 LU-decomposition Basic algorithm Cost Stability Pivoting Cholesky decomposition Sparse matrices and reorderings Gerard Sleijpen

More information