ThError: Detection of 1000-error

In statisticsnorway/Kostra: Functions for Kostra

Detection of 1000-error

Description

Detects units with possible 1000-error by comparing values in period t with revised values from period t-1

Usage

ThError(data, id, x1, x2, ll = -2.5, ul = 2.5)

Arguments

data	Input data set of class data.frame.
id	Name of identification variable.
x1	Name of variable in period t.
x2	Name of variable in period t-1.
ll	Lower limit of log10(x1/x2) = log10(x1) - log10(x2). Default -2.5
ul	Upper limit of log10(x1/x2) = log10(x1) - log10(x2). Default +2.5

Value

Output of ThError is a data set of class data.frame. Thousand Error is only checked if both x1 and x2 are not missing and not zero. The variables are:

id	The input identification variable.
x1	The input x1 variable
x2	The input x2 variable
outlier	A binary (1/0) variable indicating whether the we suspect a 1000 error or not
diffLog10	The difference log10(x1) - log10(x2)
lowerLimit	The input parameter ll
upperLimit	The input parameter ul

Author(s)

Magnar Lillegård

References

Datarevisjon: Kontroll, granskning og retting av data. Anbefalt praksis, Statistisk sentralbyrås håndbøker, 2005/84.

Examples

 minedata <- KostraData("testdata")
 minedata$areal_381_eier_2015[c(1, 4, 5)] <- 1000 * minedata$areal_381_eier_2015[c(1, 4, 5)]
 ThError(data = minedata, id = "Region", x1 = "areal_381_eier_2015", x2 = "areal_381_eier_2014")
 ThError(data = minedata, id = "Region", x1 = "areal_381_eier_2015", x2 = "areal_381_eier_2014",
         ll = -2, ul = 2)

statisticsnorway/Kostra documentation built on Nov. 2, 2024, 6:40 p.m.