[R] Creating NA equivalent

[John Dougherty]

On Tue, 21 Dec 2021 05:41:31 +0100
Marc Girondot via R-help <r-help using r-project.org> wrote:

> Dear members,
> 
> I work about dosage and some values are bellow the detection limit. I 
> would like create new "numbers" like LDL (to represent lower than 
> detection limit) and UDL (upper the detection limit) that behave like 
> NA, with the possibility to test them using for example is.LDL() or 
> is.UDL().
> 
> Note that NA is not the same than LDL or UDL: NA represent missing
> data. Here the data is available as LDL or UDL.
> 
> NA is built in R language very deep... any option to create new
> version of NA-equivalent ?
> 
> Thanks
> 
> Marc

You are concerned with a distinct quality in the data with respect to a
specific method. You might want to code a qualitative variable that
defines the detectability state of the specific reading.  Then filter on
the state of interest, and as a means of establishing the quality of the
method or the data, summarize the detection properties in your sample
for the anaytical method employed. I had an engineer tell me flatly
that the measures claimed in a paper were "impossible."  The method
used was already common, but his system was not sensitive enough.  

As far as the statistical properties go, there are measures that could
be made and measures that could not be made.  If a different method
became available, you would probably still desire to either reanalyze
the older data employing the new method, or append new measures where
they were previously unavailable. Either way you encounter
data range or compatibility issues that have to be addressed
methodologically.