Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. It only takes a minute to sign up.
Sign up to join this community
I am using R and Amelia to impute missing data for the number of homeless children in several locations. There is information about the TOTAL number of homeless children across all locations, but many individual location stats are not available. Can I create a “constraint” in my imputation model that ensures that the resulting/imputed total of the distribution is not greater than the known total.
Also, recommendations on what imputation primer to read are welcome!
To make sure your imputed total does not exceed the actual total, you could just perform multiple imputations with the m option in amelia() and choose one that meets this criteria.
Just to clarify, are you including other data for each location besides the number of homeless children? Your imputation will not be meaningful if you do not incorporate other data about each location, as missing data will be imputed based on similarity/dissimilarity of the locations based on the non-missing data for each location.
