CODEWITHSUNDEEP
excelmatlab
Bazman
Bazman

Reputation: 2150

Reading mix between numeric and non-numeric data from excel into Matlab

I have a matrix where the first column contains dates and the first row contains maturities which are alpha/numeric (e.g. 16year).

The rest of the cells contain the rates for each day, which are double precision numbers.

Now I believe xlsread() can only handle numeric data so I think I will need something else or a combination of functions?

I would like to be able to read the table from excel into MATLAB as one array or perhaps a struct() so that I can keep all the data together.

The other problem is that some of the rates are given as '#N/A'. I want the cells where these values are stored to be kept but would like to change the value to blank=" ". What is the best way to do this? Can it be done as part of the input process?

Upvotes: 0

Views: 4238

Answers (1)

Michael
Michael

Reputation: 379

Well, from looking at matlab reference for xlsread you can use the format

[num,txt,raw] = xlsread(FILENAME)

and then you will have in num a matrix of your data, in txt the unreadable data, i.e. your text headers, and in raw you will have all of your data unprocessed. (including the text headers).

So I guess you could use the raw array, or a combination of the num and txt.

For your other problem, if your rates are 'pulled' from some other source, you can use

=IFERROR(RATE DATA,"")

and then there will be a blank instead of the error code #N\A.

Another solution (only for Windows) would be to use xlsread() format which allows running a function on your imported data,

[num,txt,raw,custom] = xlsread(filename,sheet,xlRange,'',functionHandler)

and let the function replace the NaN values with blank spots. (and you will have your output in the custom array)

Upvotes: 1

Related Questions