Roaster VII/VIII, Bandana Castor, and Effiou [🇷 for BE/BA]

posted by ElMaestro  – Denmark, 2024-04-04 14:30 (285 d 14:01 ago) – Posting: # 23934
Views: 5,697

Hi Hötzi,


❝ Try this one:


   model      r.sq  r.sq.adj

 Pearson 0.6732094        NA

  simple 0.4532110 0.4415771

  nested 0.8685853 0.8598244


This would amount to a perfect corr. in other algorithms. Using an approach not unlike jackkniving you can simply do a "shift N time points, then correlate" and iterate over N, or use a subset of N consecutive points from one vector and correlate with N consecutive points from the other vector while starting at position a in vector 1 and starting at position b in the other vector. You'd then iterate over N, a and b. In R, these things take a lot of time, but in C it is really fast. Like a speed gain of a factor 10000 etc.
Detection algos called Roaster VII/VIII, Bandana Castor, and Effiou etc would take care of these cases and a lot of other (possibly imaginary?) stuff, too. No publication detailing these approaches yet; might do it if or when the time is right.

Pass or fail!
ElMaestro

Complete thread:

UA Flag
Activity
 Admin contact
23,362 posts in 4,906 threads, 1,675 registered users;
108 visitors (0 registered, 108 guests [including 18 identified bots]).
Forum time: 03:31 CET (Europe/Vienna)

I have not failed 700 times. I have not failed once.
I have succeeded in proving
that those 700 ways will not work.    Thomas Alva Edison

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz
HTML5