Adams-Bashforth and Adam-Moulton

  • Thread starter Thread starter johnchau123
  • Start date Start date
Click For Summary
SUMMARY

The discussion focuses on the use of the Adams-Bashforth method as a predictor and the Adams-Moulton method as a corrector in numerical integration. It is established that after calculating y_n+1 using the corrector method, one must recalculate y'_n+1 for the subsequent step y_n+2, rather than relying on the original value obtained from the predictor method. This ensures accuracy in the integration process, particularly when the function f(y,t) is dependent on y.

PREREQUISITES
  • Understanding of numerical integration methods, specifically Adams-Bashforth and Adams-Moulton.
  • Familiarity with differential equations and the notation dy/dt = f(y,t).
  • Knowledge of predictor-corrector algorithms in numerical analysis.
  • Basic proficiency in programming or computational tools for implementing these methods.
NEXT STEPS
  • Study the implementation of the Adams-Bashforth method in Python using libraries like NumPy.
  • Explore the Adams-Moulton method and its variations for different types of differential equations.
  • Learn about error analysis in numerical integration to understand the implications of using predictor-corrector methods.
  • Investigate other numerical methods for solving ordinary differential equations, such as Runge-Kutta methods.
USEFUL FOR

Mathematicians, engineers, and computer scientists involved in numerical analysis, particularly those working with differential equations and seeking to improve the accuracy of their solutions through advanced integration techniques.

johnchau123
Messages
8
Reaction score
0
Hi,
I am doing a question on using Adams-Bashforth method as predictor and Adams-Moulton method as corrector.
I would like to ask after we find y_n+1 by corrector method, do we have to re-calculate y'_n+1 in order to find y_n+2, or we use the y'_n+1 originally found by the predicator method to find y_n+2 in corrector?

Thanks.
:smile:
 
Physics news on Phys.org
Yes you do.

When you to the second integration step, you want to use "the answer" from the first step, not an earlier guess at the answer.

Of course if it happens that in your equation dy/dt = f(y,t), f only depends of t and not on y, there may be no difference between the two values.
 

Similar threads

Graduate Where Can Adams-Bashforth-Moulton Method be Applied in Real-World Sciences?
  • · Replies 5 ·
Replies
5
Views
4K
High School Euler's method for second order DE
  • · Replies 5 ·
Replies
5
Views
2K
Differential Equation - Where am I going wrong?
  • · Replies 2 ·
Replies
2
Views
3K
Discretize using a forward-Euler scheme
  • · Replies 1 ·
Replies
1
Views
2K
How Does Bayesian Estimation Determine Point Estimates with Prior Distributions?
  • · Replies 3 ·
Replies
3
Views
3K
A predictor-corrector method and stability
  • · Replies 1 ·
Replies
1
Views
2K
Second Order Runge Kutta for Simple Harmonic Motion
  • · Replies 1 ·
Replies
1
Views
3K
Question about approximate numerical integration methods
  • · Replies 4 ·
Replies
4
Views
1K
Solving differential equation (t^2-1)y''-6y=1
  • · Replies 2 ·
Replies
2
Views
1K
Discrete Optimization Question
  • · Replies 10 ·
Replies
10
Views
5K