Re: Math128
Soroosh Yazdani
looking at estimated integrals that you get in part a, or you can
prove the recurrence by integrating by parts. I prefer the second method,
and you can have that in your explanation. Don't worry about plotting anything
for this problem.
For part (d), we are not asking you to measure stability, rather just
give me a qualitative explanation on weather the method is stable or
unstable. Similarly, for accuracy and efficiency, just give some qualitative
comment (i.e. method b is efficient, since it is just few multiplication...)
The comment that the "more function evaluation, the more accurate the
approximation is" is not strictly true. This is the case for certain methods,
such as trapazoid, Simpsons, and Romberg integration. However, this is
probably for Newton-Cotes methods.
Soroosh
On Tue, Oct 31, 2006 at 09:23:38AM -0800, lovely0...@berkeley.edu wrote:
> Hi Mr. Yazdani,
>
> My name is Ziping Lin from your math128A. I have some questions regarding
> the programming assignment. For the Problem 2 part (b), how can I verify
> that the integrals defined in (a) satisfy the recurrence? Should I plot
> the numbers into the recurrence formula? And also for part (d),how to
> measure the stability? Finally, for the explaination, is it true that the
> more function evaluation, the more accurate the approximation is? but what
> is the specific reason for that? Could you please reply my email within
> today? If it is too long, do you have extra office hour today for me to
> drop in? Thank you very much.
>
> Ziping Lin