Hello Genie community,
I have already successfully discretized my continuous variables to build the Bayesian models, but I was wondering if the boundaries were inclusive (<= or >=) or exclusive (< or >). So if I have age as a variables that ranges from 35-100 and the bin breakpoints are at 60 and 80, so <35, 35-60, 60-80, 80-100, >100, then when are the boundaries inclusive and when are they exclusive? I might be starring at it and just can't see it, but I can't tell with the discretization interface.
Any insight would be appreciated!
Thanks!
Discretization boundaries (inclusive/exclusive)
-
- Site Admin
- Posts: 1419
- Joined: Mon Nov 26, 2007 5:51 pm
Re: Discretization boundaries (inclusive/exclusive)
The data discretization boundaries are inclusive on the left side and exclusive on the right side. The effective intervals for your example would be:
(-inf, 35) [35, 60) [60, 80) [80, 100) [100, +inf)
(-inf, 35) [35, 60) [60, 80) [80, 100) [100, +inf)
Re: Discretization boundaries (inclusive/exclusive)
Great, thank you so much!