(using taylor polynomials) Calculate ln(2.5) to 7 correct digits. (hint:use ln(e)=1, e is about 2.718)
expand about \(e\)?
\(f(3)=1\) is the constant, then \(f'(x)=\frac{1}{x}\) so \(f'(e)=\frac{1}{e}\) we start with \[f(x)=1+\frac{1}{e}(x-e)\]
thats what i am thinkin x=e, but i dont know if my answer is correct. My prof, dosent believe in giving out the answers. even to check work
next term would be \[f''(x)=-\frac{1}{x}\] so \[f''(e)=\frac{1}{e^2}\] we get \[1+\frac{1}{e}(x-e)-\frac{1}{2e^2}(x-e)^2+...\]
if you ask me it is kind of silly since you cannot do this without a calculator anyway. might as well just find \(\ln(2.5)\)
can i ask a question? why do i need to approximate ln 2.5 if i get a number when i plug it into a calc?
like i said, this is kind of silly you cannot compute \(\frac{1}{e}\) without a calculator lets try an experiment
do you know something about graphing the remainder of a taylor poylnomial?
no
but you would say that is the right approach right? evalauting the taylor poylnomial at e?
you wouldnt have by any chance taken numerical anaylsis?
http://www.wolframalpha.com/input/?i=1%2B%281%2Fe%29%28-.218%29-%281%2F%282e^2%29%29%28-.218%29^2 http://www.wolframalpha.com/input/?i=ln%282.5%29
i think we got close i am pretty sure that is what was in mind for this problem
we would have to carry the expansion out further to get closer
how many terms do you think?
not sure, but the derivatives are easy enough, so i will be ok to do a couple more
cool, thanks man.
\[\ln(x)\] \[\frac{1}{x}\] \[-\frac{1}{x^2}\] \[\frac{2}{x^3}\] \[-\frac{6}{x^4}\] for the first ones so \[1,\frac{1}{e},-\frac{1}{e^2},\frac{2}{e^3},-\frac{6}{e^4}\] for the successive derivatives
i think i got it. it looks like it weill take 6 terms
\[1+\frac{1}{e}(x-e)-\frac{1}{2e^2}(x-e)^2+\frac{1}{3e^3}(x-e)^3-\frac{1}{4e^4}(x-e)^4\] i think is right
doing that gives us 0.9162907
comparing that to the ln 2.5 i think that 7 digits do match up
but whats giving me fits is that i have to use the taylor remainder to prove that i have accuracy up to that 7th digit
pattern becomes clear you get yeah looks like it .9162907318 according to wolfram
what is the bound on the remainder?
isn't it just the next term? or am i mistaken?
yeah its, k+1
so if we take it out this far, next one would be \[\frac{1}{5e^5}(2.5-e)^5\]
as long as this is close enough to zero, you are fine
i guess so. Thanks for your help. You wouldnt be familair with maple would you?
Join our real-time social learning platform and learn together with your friends!