Identifies outliers using the algorithm lookout.

This function identifies outliers using the algorithm lookout, an outlier detection method that uses leave-one-out kernel density estimates and generalized Pareto distributions to find outliers.

lookout(X, alpha = 0.05, unitize = TRUE, bw = NULL, gpd = NULL, fast = TRUE)

Arguments

X

The input data in a dataframe, matrix or tibble format.

alpha

The level of significance. Default is 0.05.

unitize

An option to normalize the data. Default is TRUE, which normalizes each column to [0,1].

bw

Bandwidth parameter. Default is NULL as the bandwidth is found using Persistent Homology.

gpd

Generalized Pareto distribution parameters. If `NULL` (the default), these are estimated from the data.

fast

If set to TRUE, makes the computation faster by sub-setting the data for the bandwidth calculation.

Value

A list with the following components:

outliers

The set of outliers.

outlier_probability

The GPD probability of the data.

outlier_scores

The outlier scores of the data.

bandwidth

The bandwdith selected using persistent homology.

kde

The kernel density estimate values.

lookde

The leave-one-out kde values.

gpd

The fitted GPD parameters.

Examples

X <- rbind(
  data.frame(x = rnorm(500),
             y = rnorm(500)),
  data.frame(x = rnorm(5, mean = 10, sd = 0.2),
             y = rnorm(5, mean = 10, sd = 0.2))
)
lo <- lookout(X)
lo
#> Leave-out-out KDE outliers using lookout algorithm
#> 
#> Call: lookout(X = X)
#> 
#>   Outliers Probability
#> 1      501  0.02261325
#> 2      502  0.02276128
#> 3      503  0.02294858
#> 4      504  0.02299057
#> 5      505  0.02289632
#> 
autoplot(lo)