Skip to contents

Evaluate Event Independence

mm_independence.Rd

This function calculates the difference between times and evaluates whether events are independent based on a given threshold. It is useful for checking if certain events in a dataset are independent of each other based on time intervals.

Usage

mm_independence(
  data = NULL,
  species_column = NULL,
  datetime,
  format,
  threshold = 30 * 60,
  only = FALSE
)

Arguments

data

A data.frame, tbl_df, or tbl containing the event data. This should include a column with datetime values. If NULL, the function will use the deltatime argument instead of the data argument.

species_column

An optional column name specifying the species grouping. If provided, independence will be assessed separately within each species group.

datetime

A character string specifying the name of the column in data that contains the datetime values. This argument is required if data is provided.

format

A character string defining the format used to parse the datetime values in the datetime column.

threshold

A numeric value representing the time difference threshold (in seconds) to determine whether events are independent. Events are considered independent if the time difference between them is greater than or equal to this threshold. The default is 30 minutes (1800 seconds).

only

A logical value indicating whether to return only the rows of data that are identified as independent events. If TRUE, only independent events are returned. If FALSE, the entire data frame is returned with an additional column indicating the independence status. The default is TRUE.

Value

  • If data is provided and only is TRUE, a tibble of events identified as independent.

  • If data is provided and only is FALSE, a tibble of the original data with additional columns indicating the independent status and deltatime differences.

  • If data is not provided, a tibble of the deltatime values with independent status.

Examples


library(dplyr)
#> 
#> Attaching package: ‘dplyr’
#> The following objects are masked from ‘package:stats’:
#> 
#>     filter, lag
#> The following objects are masked from ‘package:base’:
#> 
#>     intersect, setdiff, setequal, union

# Load example dataset
cam_data <- read.csv(system.file("penessoulou_season1.csv", package = "maimer"))

# Independence without considering species occurrence
indep1 <- cam_data %>%
  mm_independence(data = ., datetime = datetimes, format = "%Y-%m-%d %H:%M:%S",
                  only = TRUE)

sprintf("Independent observations: %s", nrow(indep1))
#> [1] "Independent observations: 177"

# Independence considering species occurrence
indep2 <- cam_data %>%
  mm_independence(data = ., datetime = datetimes, format = "%Y-%m-%d %H:%M:%S",
                  only = TRUE, species_column = "species")

sprintf("Independent observations: %s", nrow(indep2))
#> [1] "Independent observations: 183"

# Use a standalone vector of datetime values
dtime <- cam_data$datetimes
mm_independence(datetime = dtime, format = "%Y-%m-%d %H:%M:%S", only = TRUE)
#> # A tibble: 177 × 1
#>    datetime           
#>    <dttm>             
#>  1 2019-01-20 15:24:41
#>  2 2019-01-21 15:23:38
#>  3 2019-01-23 15:35:20
#>  4 2019-01-24 06:01:07
#>  5 2023-09-20 16:15:28
#>  6 2023-09-21 13:32:45
#>  7 2024-03-02 22:32:10
#>  8 2024-03-10 10:21:56
#>  9 2024-03-10 15:04:38
#> 10 2024-03-10 17:28:11
#> # ℹ 167 more rows