[R] fft help

[Rolf Turner]

On 17/09/2009, at 3:39 AM, delic wrote:

>
> I wrote a script that I anticipating seeing a spike at  10Hz with the
> function 10* sin(2*pi*10*t).
> I can't figure out why my plots  do not show spikes at the  
> frequencies I
> expect. Am I doing something wrong or is my expectations wrong?

(a) Is this a homework question?

(b) Have you figured it out yet?

(c) Hint:  You have spikes at +/- 40 in a range from -50 to 50. You  
*want* spikes
at 10 and 90 Hz. Could it be that you haven't set your frequency  
vector ``f''
quite right? :-)

	cheers,

		Rolf Turner

P. S. You won't get spikes bang on at 10 and 90 Hz. because these are  
*not*
Fourier frequencies when n = 256.  If you want spikes in your  
periodogram
at bang on 10 and 90 Hz use a value of n that is divisible by 10,  
e.g. n=500.
Why would you want a power of 2 anyhow?  (Well, the fft goes faster  
when n
is a power of 2, but who cares?)

		R. T.

######################################################################
Attention:\ This e-mail message is privileged and confid...{{dropped:9}}