Do we fit data with, by, or as a linear function?
|
None of those: we fit data to a function. See the usage throughout Wolfram Alpha, for example. |
|||||||||
|
|
I would say you fit a function or model to the data, rather than the other way around. Or you find a model that fits the data. |
|||
|
|
It's kind of weird for me to say "fit data to a function". I don't think there is a big difference either way, but I prefer "fit a function to the data" and I think this latter way is the form I've always heard or read. I think of it like how a tailor "fits" a dress or suite to a person's body, where the measurements are the data and the clothing piece is the function. |
|||||||
|
|
Actually, you don't do anything to data, which is the point to note. You could say 'the data obeys/ conforms to/ or even fits, a (linear) function.' It's important to note the difference between the inherent nature of the data and actions performed on the data. |
||||
|
|