Create site with anomalous patients

Average

Usage

anomaly_average(df, anomaly_degree, site = "sample_site")

anomaly_sd(df, anomaly_degree, site = "sample_site")

anomaly_autocorr(df, anomaly_degree, site = "sample_site")

anomaly_lof(df, anomaly_degree, site = "sample_site", verbose = FALSE)

anomaly_range(df, anomaly_degree, site = "sample_site")

anomaly_unique_value_count_relative(df, anomaly_degree, site = "sample_site")

Arguments

df

Data frame containing the study data.

anomaly_degree

Degree of anomaly to add.

site

prefix for new sites, Default is "sample_site".

verbose

logical, Default: FALSE

Value

A data frame with added anomalies.

See also

rsurprise

Examples

set.seed(1)
library(ggplot2)

df_prep <- prep_sdtm_lb(pharmaversesdtm::lb, pharmaversesdtm::dm, scramble = TRUE)

df_filt <- df_prep %>%
  filter(parameter_id == "Alkaline Phosphatase")

df_anomaly <- anomaly_average(df_filt, anomaly_degree = 2, site = "anomolous")

ggplot(df_filt, aes(x = timepoint_rank, y = result, group = subject_id)) +
geom_line(color = "black") +
geom_line(data = df_anomaly, color = "tomato") +
coord_cartesian(xlim = c(0, max(df_anomaly$timepoint_rank)))


set.seed(1)
library(ggplot2)

df_prep <- prep_sdtm_lb(pharmaversesdtm::lb, pharmaversesdtm::dm, scramble = TRUE)

df_filt <- df_prep %>%
  filter(parameter_id == "Alkaline Phosphatase")

df_anomaly <- anomaly_sd(df_filt, anomaly_degree = 2, site = "anomolous")

ggplot(df_filt, aes(x = timepoint_rank, y = result, group = subject_id)) +
geom_line(color = "black") +
geom_line(data = df_anomaly, color = "tomato") +
coord_cartesian(xlim = c(0, max(df_anomaly$timepoint_rank)))


set.seed(1)
library(ggplot2)

df_prep <- prep_sdtm_lb(pharmaversesdtm::lb, pharmaversesdtm::dm, scramble = TRUE)

df_filt <- df_prep %>%
  filter(parameter_id == "Alkaline Phosphatase")

df_anomaly <- anomaly_autocorr(df_filt, anomaly_degree = 2, site = "anomolous")

ggplot(df_filt, aes(x = timepoint_rank, y = result, group = subject_id)) +
geom_line(color = "black") +
geom_line(data = df_anomaly, color = "tomato") +
coord_cartesian(xlim = c(0, max(df_anomaly$timepoint_rank)))


set.seed(1)
library(ggplot2)

df_prep <- prep_sdtm_lb(pharmaversesdtm::lb, pharmaversesdtm::dm, scramble = TRUE)

df_filt <- df_prep %>%
  filter(parameter_id == "Alkaline Phosphatase")

df_anomaly <- anomaly_lof(df_filt, anomaly_degree = 2, site = "anomolous", verbose = TRUE)
#> Selected distribution: runif 
#> Parameters: min = 1.7847; max = 9.6711 
#> Selected distribution: rpois 
#> Parameters: lambda = 1.3228 
#> Selected distribution: rgamma 
#> Parameters: shape = 8.7425; scale = 1.9169 
#> Selected distribution: rpois 
#> Parameters: lambda = 7.688 
#> Selected distribution: rpois 
#> Parameters: lambda = 1.312 
#> Selected distribution: rpois 
#> Parameters: lambda = 1.6909 
#> Selected distribution: rexp 
#> Parameters: rate = 1.8951 
#> Selected distribution: rexp 
#> Parameters: rate = 0.3266 
#> Selected distribution: rgamma 
#> Parameters: shape = 1.9628; scale = 0.5289 
#> Selected distribution: rbinom 
#> Parameters: size = 43; prob = 0.9261 
#> Selected distribution: rpois 
#> Parameters: lambda = 7.7639 
#> Selected distribution: rpois 
#> Parameters: lambda = 2.8724 
#> Selected distribution: rexp 
#> Parameters: rate = 1.3202 
#> Selected distribution: rbeta 
#> Parameters: shape1 = 1.0197; shape2 = 2.4954 
#> Selected distribution: runif 
#> Parameters: min = 4.9997; max = 7.6387 
#> Selected distribution: runif 
#> Parameters: min = 2.654; max = 7.685 
#> Selected distribution: rgamma 
#> Parameters: shape = 6.579; scale = 1.8946 
#> Selected distribution: rpois 
#> Parameters: lambda = 3.3932 
#> Selected distribution: rgamma 
#> Parameters: shape = 1.7505; scale = 1.5516 
#> Selected distribution: rbeta 
#> Parameters: shape1 = 0.6712; shape2 = 4.6334 
#> Selected distribution: rbeta 
#> Parameters: shape1 = 0.2613; shape2 = 3.5113 
#> Selected distribution: rpois 
#> Parameters: lambda = 9.4547 

ggplot(df_filt, aes(x = timepoint_rank, y = result, group = subject_id)) +
geom_line(color = "black") +
geom_line(data = df_anomaly, color = "tomato") +
coord_cartesian(xlim = c(0, max(df_anomaly$timepoint_rank)))



set.seed(7)
library(ggplot2)

df_prep <- prep_sdtm_lb(pharmaversesdtm::lb, pharmaversesdtm::dm, scramble = TRUE)

df_filt <- df_prep %>%
  filter(parameter_id == "Alkaline Phosphatase")

df_anomaly <- anomaly_range(df_filt, anomaly_degree = 2, site = "anomolous")

ggplot(df_filt, aes(x = timepoint_rank, y = result, group = subject_id)) +
geom_line(color = "black") +
geom_line(data = df_anomaly, color = "tomato") +
coord_cartesian(xlim = c(0, max(df_anomaly$timepoint_rank)))


set.seed(1)
library(ggplot2)

df_prep <- prep_sdtm_lb(pharmaversesdtm::lb, pharmaversesdtm::dm, scramble = TRUE)

df_filt <- df_prep %>%
  filter(parameter_id == "Alkaline Phosphatase")

df_anomaly <- anomaly_unique_value_count_relative(df_filt, anomaly_degree = 2, site = "anomolous")

ggplot(df_filt, aes(x = timepoint_rank, y = result, group = subject_id)) +
geom_line(color = "black") +
geom_line(data = df_anomaly, color = "tomato") +
coord_cartesian(xlim = c(0, max(df_anomaly$timepoint_rank)))