Single formula to return multiple column averages?

English
#1

I have a felling there is no way around writing a formula for each column group) rather than one formula that would return multiple values, but

I am working ion color analysis and whether RGB or Lab*, there will always be 3 values per color. I have blocks of data of varying size that I need the averages which become the centroid values to generate ∆E (didtsnver between each data set and the centroid set.

Just curious if there might be a single formula to accomplish this?

Thanks in advance (and I hope this isn’t too sill of a question)

Stan

#2

Average takes up to 30 arguments. =AVERAGE(B1:B99;D1:D99;F1:F99)

#3

I tried this.=AVERAGE(A22:A25,B22:B25,C22:C25). I used semi-colons, but it changes them to commas, and it appears it returns a sum total average, raster than 3 sets of values

#4

Please attach an example of what you expect to be returned, using this button:
image
because your request is not clear for me. If you want to return 3 values, then it is easier to have three formulas in three different cells in my opinion.

#6

(post deleted by author)

#7

RGB has HEX format, thus 256×256×256 (00…FF) constellations!

#9

Steph1. I am beginning to agree with you!

#5

… in CMYK you will define 4 basic “colours” in a wider spectrum as RGB. Do you want to analyse the reduced size of a cheap monitor or a special one with high color fidelity or the logarithmic sensory system of human perceptions of 18 billion different eyes in variations of the black and white seeing highly sensitive night light photoreceptors (contrast) to the high luminosity perceiving day light photoreceptors (color)?
In all there are no linear skalings! So what and/or witch you want to get a average value of what and/or why?
I think you’re trying to simplify an extremely complex topic somehow? I’m hoping for a system that takes into account people with different color orientations, those with reduced/increased red and/or green and/or blue and/or orange vision. The topic is anything but simple. I did not know anything about color average formulas!

#8

Koyotak. I am analyzing Lab* data acquired from raw data using flatbed scanner (lots of documents uses for this, Scatnner is IT8 calibrated and monitors are HD on a new Mac OSX15 system. I am studying the inks used in printing late 1800/early 1900 postage stamps. Lab* is my preference. The averaging is of clusters for centroid so I can plot spheres with a radii = to 2 ∆E. Using the K-means functions a starting point but it does not handle outliers well. the 3D scatter app is python based, which I built myself as I could not find anything that specifically addressed what I get were my needs.

I am retired, so if it turns out for naught, it was at least a mostly enjoyable use of my time.

Cheers

Impressum (Legal Info) | Datenschutzerklärung (Privacy Policy)   Content on this site is licensed under a Creative Commons Attribution Share Alike 3.0 license.