Skip to content


Repository files navigation

R build status CRAN Downloads status Codecov test coverage Mentioned in Awesome Official Statistics

Error localization

Find errors in data given a set of validation rules. The errorlocate helps to identify obvious errors in raw datasets.

It works in tandem with the package validate. With validate you formulate data validation rules to which the data must comply.

For example:

  • “age cannot be negative”: age >= 0.
  • “if a person is married, he must be older then 16 years”: if (married ==TRUE) age > 16.
  • “Profit is turnover minus cost”: profit == turnover - cost.

While validate can check if a record is valid or not, it does not identify which of the variables are responsible for the invalidation. This may seem a simple task, but is actually quite tricky: a set of validation rules forms a web of dependent variables: changing the value of an invalid record to repair for rule 1, may invalidate the record for rule 2.

errorlocate provides a small framework for record based error detection and implements the Felligi Holt algorithm. This algorithm assumes there is no other information available then the values of a record and a set of validation rules. The algorithm minimizes the (weighted) number of values that need to be adjusted to remove the invalidation.


errorlocate can be installed from CRAN:


Beta versions can be installed with drat:


The latest development version of errorlocate can be installed from github with devtools:



#> Loading required package: validate
rules <- validator( profit == turnover - cost
                  , cost >= 0.6 * turnover
                  , turnover >= 0
                  , cost >= 0 # is implied

data <- data.frame(profit=750, cost=125, turnover=200)

data_no_error <- replace_errors(data, rules)

# faulty data was replaced with NA
#>   profit cost turnover
#> 1     NA  125      200

er <- errors_removed(data_no_error)

#> call:  locate_errors(data, x, ref, ..., cl = cl) 
#> located  1  error(s).
#> located  0  missing value(s).
#> Use 'summary', 'values', '$errors' or '$weight', to explore and retrieve the errors.

#> Variable:
#>       name errors missing
#> 1   profit      1       0
#> 2     cost      0       0
#> 3 turnover      0       0
#> Errors per record:
#>   errors records
#> 1      1       1

#>      profit  cost turnover
#> [1,]   TRUE FALSE    FALSE


No releases published


No packages published