change the resolution of time series from hourly to 30min

Hello all!

I've got some hourly data and I want to transpose it to 30min data.

For example for the variable 1, I've got the timeseries: (Dates - Values:)

00:00 01/07/2010 -> 5

01:00 01/07/2010 -> 6

02:00 01/07/2010 -> 8

and I need to transpose it to: (Dates - Values)

00:00 01/07/2010 -> 5

00:30 01/07/2010 -> 5

01:00 01/07/2010 -> 6

01:30 01/07/2010 -> 6

02:00 01/07/2010 -> 8

02:30 01/07/2010 -> 8

Do you have any idea??

Answer 1

[Fangjun Jiang]

resample() would be the function to use, but in your case, just process your original data to create a new timeseries.

Data=[5 6 8];
Time={'00:00 01/07/2010','01:00 01/07/2010','02:00 01/07/2010'};
ts=timeseries(Data',Time')

NewData=repmat(Data,2,1);
NewTime=datenum(Time);
NewTime=[NewTime';NewTime'+0.5/24];
NewTs=timeseries(NewData(:),NewTime(:))

[Christina]

Hey!

Thanks for answering.

However, with the repmat command the vector Data = [5 ;6; 8] is repeated and it becomes [5 ;6 ;8; 5 ;6 ;8]

whereas I want it to become [5; 5; 6; 6; 8; 8]

Do you know how this can be done?

[Fangjun Jiang]

In my code above, Data=[5 6 8] is a row vector, not a column vector as you just mentioned Data=[5;6;8].

If you have Data=[5;6;8], you can do NewData=[Data Data]';NewData=NewData(:); you'll get what you want.