MATLAB Timeseries Resample Error: Quality vector must be initialized.

Hi there, I am trying to resample a timeseries on MATLAB using the following code:

 load RainData;
 RKraut = RainData.Krautheim.data;
 TKraut = RainData.Krautheim.time;
 OldTS = timeseries(RKraut,TKraut);
 NewT = ceil (TKraut(1)):1/1440:floor(TKraut(end));
 NewRKraut = resample(OldTS,NewT,'interpmethod','zoh');
 Krautheim.data = NewRKraut.data;
 Krautheim.time = NewRKraut.time;

if I do not specify the interpolation method, everything works fine, but once I specify interpolation method as 'zoh', 'phcip' or 'linear' I get this error.

Error using timeseries/resample (line 65) Quality vector must be initialized.

Error in KrautheimRainResampling (line 6) NewRKraut = resample(OldTS,NewT,'interpmethod','zoh');

I am not even sure what is meant by 'Quality vector', I checked online and couldn't find any useful related subjects. I am using MATLAB R2018b. Any help would be highly appreciated.

Thanks in advance!

Answer 1

[jonas]

I suggest that you use a timetable and retime instead. I believe timetable offers more or less all the same utility as the timeseries function and more. Upload data if you need help adapting your code.

[Ghazi Eltiti]

Thank you for your suggestion. It would have worked if the retime function offered the zero-order hold interpolation method. Any idea why I am getting that error though?

[jonas]

That's new to me, personally I've never had to use anything other than linear interp. I don't know what the problem is but I can take a look if you upload the data or a sample of the data for which the error is reproduced.

If I had to guess, then I'd make sure that all your old time-steps are unique. Could be that you mess up the vector when you round.

[Ghazi Eltiti]

The data are unique, and the function works if I do not specify the method of interpolation (linear interpolation by default), however, I get this error even if I specify the method as linear. There is no need to upload the data since I found a work around. But thanks for the offering to help.