key_creation: key_creation

In 1Edtrujillo1/udeploy: Back-end & Front-end functions

Description

Create a unique key ID variable

Usage

key_creation(df, factor_vars = NULL)

Arguments

df	dataset to add a unique ID variable
factor_vars	factor variables reference

Details

This function allows you to add a unique key ID to the dataset based on the factor variables (or not)

Value

"This function allows us to return original dataset with a new KEY variable based on:"

• If factor_vars = NULL bring a KEY variable as day_month_year_sequence from 1 to the length of the dataset

• If factor_vars is not NULL bring a KEY variable as day_month_year_concatenate the unique id for each level for each factor variable selected

Note

• The intention of this function is that in the future we use this new variable as an ID for merges.

• If for some reason the key variable is not unique then the program re-ajust it, so filter those duplicated observations and concatenate them with the corresponding index number (from 1 to the length of the duplicated values).

• If the dataset has a date variable then use this variable as reference but if not is going to create a new date variable reference

• today()+(1:length(vector)) is: sequence of days from today to the day that represent the last element of the vector

• today()+(1:length(vector))*days(-1) is: days(-1) invert the order, sequence this is from the day that represent the first element of the vector to yesterday "today()-1"

• sort(today()+(1:length(df1$ID_REGION))*days(-1)) is: sorted day from the lowest to the gratest

Author(s)

Eduardo Trujillo

1Edtrujillo1/udeploy documentation built on July 13, 2021, 9:12 p.m.