Skip to contents

Detect outliers

Source: R/outliers.R
detect_outliers.Rd

It detects outliers based on prediction bands for the given level of confidence provided.

Usage

detect_outliers(
  .data,
  test_type = c("incremental", "kinetics"),
  vo2_column = "VO2",
  cleaning_level = 0.95,
  cleaning_baseline_fit,
  protocol_n_transitions,
  protocol_baseline_length,
  protocol_transition_length,
  method_incremental = c("linear", "anomaly"),
  verbose = TRUE,
  ...
)

Arguments

.data

Data retrieved from read_data() for a kinetics test, or the data retrieved from incremental_normalize() for a incremental test.

test_type

The test to be analyzed. Either 'incremental' or 'kinetics'.

vo2_column

The name (quoted) of the column containing the absolute oxygen uptake (VO2) data. Default to VO2.

cleaning_level

A numeric scalar between 0 and 1 giving the confidence level for the intervals to be calculated. Default to 0.95.

cleaning_baseline_fit

For kinetics test only. A vector of the same length as the number in protocol_n_transitions, indicating what kind of fit to perform for each baseline. Vector accepts characters either 'linear' or 'exponential'.

protocol_n_transitions

For kinetics test only. Number of transitions performed.

protocol_baseline_length

For kinetics test only. The length of the baseline (in seconds).

protocol_transition_length

For kinetics test only. The length of the transition (in seconds).

method_incremental

The method to be used in detecting outliers from the incremental test. Either 'linear' or 'anomaly'. See Details.

verbose

A boolean indicating whether messages should be printed in the console. Default to TRUE.

...

Additional arguments. Currently ignored.

Value

a tibble

Details

TODO

Examples

if (FALSE) { # \dontrun{
## get file path from example data
path_example <- system.file("example_cosmed.xlsx", package = "whippr")

## read data
df <- read_data(path = path_example, metabolic_cart = "cosmed")

## detect outliers
data_outliers <- detect_outliers(
  .data = df,
  test_type = "kinetics",
  vo2_column = "VO2",
  cleaning_level = 0.95,
  cleaning_baseline_fit = c("linear", "exponential", "exponential"),
  protocol_n_transitions = 3,
  protocol_baseline_length = 360,
  protocol_transition_length = 360,
  verbose = TRUE
 )

## get file path from example data
path_example_ramp <- system.file("ramp_cosmed.xlsx", package = "whippr")

## read data from ramp test
df_ramp <- read_data(path = path_example_ramp, metabolic_cart = "cosmed")

## normalize incremental test data
ramp_normalized <- df_ramp %>%
 incremental_normalize(
   .data = .,
   incremental_type = "ramp",
   has_baseline = TRUE,
   baseline_length = 240,
   work_rate_magic = TRUE,
   baseline_intensity = 20,
   ramp_increase = 25
 )

## detect ramp outliers
data_ramp_outliers <- detect_outliers(
  .data = ramp_normalized,
  test_type = "incremental",
  vo2_column = "VO2",
  cleaning_level = 0.95,
  method_incremental = "linear",
  verbose = TRUE
 )
} # }