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
Since many functions require frequency weights, we can store it in design.variables rather than having each function derive it from probs.
Instead of probs, perhaps, we should call it probability_weights_reserved? Similarly, frequency_weights_reserved.
The reason we're only storing probability weights is that we didn't anticipate so many Julia functions to use frequency weights. I believe the survey package in R only store probability weights, and most R functions use probability weights.
Note, we want the information about weights to be in the same data frame and not make it a property of svydesign because of performance reasons. combine function in DataFrames works a lot faster.
The text was updated successfully, but these errors were encountered:
Since many functions require frequency weights, we can store it in
design.variables
rather than having each function derive it fromprobs
.Instead of
probs
, perhaps, we should call itprobability_weights_reserved
? Similarly,frequency_weights_reserved
.The reason we're only storing probability weights is that we didn't anticipate so many Julia functions to use frequency weights. I believe the
survey
package in R only store probability weights, and most R functions use probability weights.Note, we want the information about weights to be in the same data frame and not make it a property of
svydesign
because of performance reasons.combine
function inDataFrames
works a lot faster.The text was updated successfully, but these errors were encountered: