asynchronism

[rand0m]

why does high sensitivity to asynchronism result in lower correlation estimates?

 

[bigwilly]

I remember it as, asynchronism means not synchronized if you are highly not synchronized then you will have a low correlation.

 

[prockets]

asynchronus data is created by taking a little data and them making it “longer” by filling in holes w/ values close to the hole you’re filling in (e.g. missing january price, so you use december price again). this results in low volatility = low correlation.

 

[comp_sci_kid]

so if i understand it correctly if you have a monthly sequnce of market prices and make then daily, your daily vol and corr will be lower then if the market data would be available daily

 

[prockets]

i think it’s in the context of only having monthly data for X, but daily data for Y..then you create daily data for X to match up w/ Y…now it’s asychronous w/ low vol/corr.

 

[comp_sci_kid]

prockets Wrote:
——————————————————-
> i think it’s in the context of only having monthly
> data for X, but daily data for Y..then you create
> daily data for X to match up w/ Y…now it’s
> asychronous w/ low vol/corr.
rgr that