Difference between Frequency and Weight variables (Visual Statistics)

[pvareschi]

Re: Predicting Analytics on Big Data

What is the difference between a Frequency and a Weight variable? (see page 4.5-4.6 of the course text)

[ed_sas_member]

Hi @pvareschi
A WEIGTH statement can Ne used when you already have frequencies in your dataset (in a PROC FREQ for instance).
However, the main difference in my opinion is that the value of the WEIGHT variable is not required to be an integer.
Hope this helps!
Best,

[pvareschi]

Thank you for your comments; I also wonder whethe the two variables may fulfil a different roles, in terms of their effect on a fitted model...or, at least, that seems to be the case from the wording in the course text, although I cannot understand how that is applied in practice.

[Ksharp · Accepted solution]

Both would get you the same parameter estimator, but Freq would have less confident interval, Weight would be bigger. In other words Freq would get you more accurate parameter estimator / better model.

@Rick_SAS  wrote a blog about it a couple of years ago .

[Rick_SAS]

See the article "The difference between frequencies and weights in regression analysis"