Hi there
We have 25m XY pairs to be correlated.
Data is bank financial data and smooth over Time ; observations for x and y
are 32 quarters each.
Testing 25m rships exhaustively will take forever ; this task is easily
over-engineered.
We'll use the best of XY relationships to predict. *
We're swamped by choice in R packages. None seem to be readily comparable.
Can someone help us with which package(s) is/are most apt ?
That is, which can time-efficiently test for correlation given properties of
the data ?
We'll save lots of time with fine advice.
(And yes, to conserve time, we'll employ foreach & doSNOW.)
Steve
* X and Y are series both 50 quarters in length.
Y is offset 8 quarters forward, such that X (periods 9 to 50) is compared to
Y (1 to 42).
Best fit (x actual vs x fitted using y actual) is determined.
Method identified is used with Y (43 to 50) to predict X for the final 8
quarters.
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
