Flag_From_Trend: Flag values as outliers based on error from estimated trend...

In Covid19Wastewater: Prepare, Analyze, and Visualize Covid-19 Wastewater Data

Flag values as outliers based on error from estimated trend This function can be done within group if the data fed into it was grouped

Description

Flag values as outliers based on error from estimated trend This function can be done within group if the data fed into it was grouped

Usage

Flag_From_Trend(DF, base_data, trend_data, sd_degree = 2.5)

Arguments

DF	Dataframe containing selected columns
base_data	Name of column containing Raw data
trend_data	Name of generated trend data
sd_degree	Threshold to flag deviant values as an outlier

Value

DF with a new column 'flagged_outlier' that contains if the column is an outlier

Examples

library(dplyr)
data("Example_data", package = "Covid19Wastewater")
smoothing_df <- Example_data%>%
    select(site, date, N1, N2)%>%
    filter(N1 != 0, N2 != 0)%>%
    mutate(N1 = log(N1), N2 = log(N2), N12_avg = (N1 + N2) / 2)
    df_data <- loessSmoothMod(smoothing_df, "N12_avg", "N12_avg_loess", Filter = NULL)
head(df_data%>%
   group_by(site)%>%
   Flag_From_Trend( N12_avg, N12_avg_loess)%>%
   select(site, date, N12_avg, flagged_outlier))

Covid19Wastewater documentation built on Aug. 25, 2023, 1:07 a.m.