You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In R survey, population proportions and population total estimation is done by passing a 'factor' (CategoricalArray) variable to svymeansvytotal (see pg 21 Lumley textbook).
Basic proportion and count estimation functionality was added to SimpleRandomSample and started for StratifiedSample, inside their respective svymean and svytotal functions to emulate functionality from R. This is currently done using an if statement
I am thinking that it would be neater and cleaner to have these features as multiple dispatch, instead of being caught by the if. Minor improvements in speed would also be there for very large datasets by using multiple dispatch.
The text was updated successfully, but these errors were encountered:
In R
survey
, population proportions and population total estimation is done by passing a 'factor' (CategoricalArray) variable tosvymean
svytotal
(see pg 21 Lumley textbook).Basic proportion and count estimation functionality was added to
SimpleRandomSample
and started forStratifiedSample
, inside their respective svymean and svytotal functions to emulate functionality from R. This is currently done using anif
statementI am thinking that it would be neater and cleaner to have these features as multiple dispatch, instead of being caught by the
if
. Minor improvements in speed would also be there for very large datasets by using multiple dispatch.The text was updated successfully, but these errors were encountered: