Skip to contents

Flag_Poisson

Source: R/Flag_Poisson.R
Flag_Poisson.Rd

[Stable]

Add columns flagging sites that represent possible statistical outliers when the Poisson statistical method is used.

Usage

Flag_Poisson(dfAnalyzed, vThreshold = NULL)

Arguments

dfAnalyzed

data.frame where flags should be added.

vThreshold

Vector of 4 numeric values representing lower and upper threshold values. All values in the Score column are compared to vThreshold using strict comparisons. Values less than the lower thresholds or greater than the upper thresholds are flagged. Values equal to the threshold values are set to 0 (i.e., not flagged). If NA is provided for either threshold value, it is ignored and no values are flagged based on the thresholds. NA and NaN values in Score are given NA flag values.

Value

data.frame with one row per site with columns: GroupID, Numerator, Denominator, Metric, Score, PredictedCount, Flag

Details

This function flags sites based on the Poisson analysis result as part of vignette("DataModel").

Data Specification

Flag_Poisson is designed to support the input data (dfAnalyzed) generated from the Analyze_Poisson function. At a minimum, the input must define a dfAnalyzed data frame with a Score variable included and a vThreshold. These inputs will be used to identify possible statistical outliers in a new Flag column by comparing Score values to the specified thresholds.

The following columns are considered required:

  • GroupID - Group ID; default is SiteID

  • GroupLevel - Group Type

  • Score - Site residuals calculated from the rates of exposure provided to Analyze_Poisson()

Examples

dfTransformed <- Transform_Rate(analyticsInput)

dfAnalyzed <- Analyze_Poisson(dfTransformed)
#> ℹ Fitting log-linked Poisson generalized linear model of [ Numerator ] ~ [ log( Denominator ) ].

dfFlagged <- Flag_Poisson(dfAnalyzed, vThreshold = c(-7, -5, 5, 7))