Can anyone provide a simple explanation why halving the step size tends to decrease the numerical error in Euler's method by one-half?

I've looked at some online sources but they do provide very complex explanations.


Here's an AoPS thread with some fairly simple explanations.

In short: the error added in one step of length $h$ is $h^2\frac{y''(c)}{2}$, where $c$ is some point in the interval we stepped over. To cover a certain fixed distance $A$, we need about $\frac{A}{h}$ steps, so the total error looks like a constant times $h$.

There's also compounded error - in later steps, our $y$ values will be in error, and that will affect our estimated $y'$ values. With some calculation, we can show that this basically only makes the proportionality constant bigger; the error will still be proportional to $h$.

| cite | improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.