Home »
Public Forums »
archive »
Fourier transform with missing data.
| Fourier transform with missing data. [message #1620] |
Thu, 13 January 1994 10:31 |
rfinch
Messages: 51
Registered: March 1991
|
Member |
|
|
Followup-To: sci.math.num-analysis
We are doing Fourier transforms of hydrologic timeseries data and a
question occurred to us. This data often has missing values. Right
now we just remove the missing data and create a new 1-d array with
only the good data; this means there will be abrupt jumps in the data.
Then we pass the new vector to the FFT routine.
Is this the right way to handle missing data? Should they perhaps be
zeroed out, instead of removed? I think of this as an alternative
only because some textbooks advise padding your vector with zero's to
get the vector length to a power of 2.
--
Ralph Finch 916-653-8268 voice
rfinch@venice.water.ca.gov 916-653-6077 fax
Any opinions expressed are my own; they do not represent the DWR
|
|
|
|
Current Time: Wed Oct 08 15:33:45 PDT 2025
Total time taken to generate the page: 0.00528 seconds
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
