Dear Bertrand,
No, there are no other labels that are treated as missing values. You're right
that #N/A was missing from the documentation.
There is no valid numeric value that is treated as missing (there are many
different conventions, we used -9999.99 a long time ago, but that has been
abolished). You can replace specific values in your code as follows:
x = loadmat(InputFile,1);
x = x .== -9999.99 .? .NaN .: x;
Jurgen
oxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxox
oxoxoxox US OxMetrics user conference 2010
oxoxoxox George Washington University, Washington DC
oxoxoxox 18-19 March 2010
oxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxox
Dr Jurgen A Doornik
University of Oxford, Nuffield College, Oxford OX1 1NF
tel. UK: +44-1865-278610 fax +44-1865-278621
http://www.doornik.com
http://www.oxmetrics.net
oxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxoxox
Bertrand Berlemont wrote:
> Dear all,
>
>
>
> I am using version 5.1 of Ox and I am trying to load tab-separated text
> files in Ox.
>
> I am using the command “loadmat(InputFile,1);” followed by a ‘shape’
> command to reshape my data and it works fine except for the missing data.
>
> Unfortunately, the only possibilities I have to indicate missing values
> in the text file are:
>
> ‘NaN’ or any fixed double (such as 9999.9999, for instance).
>
>
>
> In Ox documentation about ‘loadmat’, I see that ‘loadmat’ is supposed to
> recognize only ‘m’, ‘M’ and .NaN as missing values but when I try it, I
> see that it also recognizes Excel’s ‘#N/A’.
>
>
>
> My question is the following: what other undocumented item is also
> regarded as a missing value by ‘loadmat’ ? In particular, is there any
> numerical value (such as 999999) that is regarded as a missing value ?
>
>
>
>
>
> Kind regards,
>
>
>
> Bertrand BERLEMONT
>
|