Re: [ojAlgo-user] Minimum Size of CalendarDateSeries when building covariance matrix

[Anders P. <an...@op...>]

Yes, for the calculations to be at all possible you need at least 3 elements in each series. (I would recommend having much more than that.) /Anders



> On 15 Feb 2017, at 23:10, Vincent de CHACUS C. <dec...@ya...> wrote:
> 
> Hey Anders,
> 
> I was running some simulations and one of my covariance matrices was returning NaN everywhere.
> I was able to trace the issue back to one of my entries, which only has 2 elements in its CalendarDateSeries (compared to an average of 10 for all the other elements in the collection).
> When I add that calendarDateSeries to the collection that serves as parameter of the makeCovarianceMatrix in the FinanceUtils class, I get a covariance matrix that's made up of NaNs.
> 
> Is there a constraint on the size of each CalendarDateSeries that make up the timeSeriesCollection?
> I'm using ojAlgo-40.0.4, although I was able to replicate the issue on version 42.0.0 of the jar.
> 
> Thanks,
> Vince 
> 
>  
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, SlashDot.org! http://sdm.link/slashdot_______________________________________________
> ojAlgo-user mailing list
> ojA...@li...
> https://lists.sourceforge.net/lists/listinfo/ojalgo-user