Calculate Predicted Bounds for Analyze_TimeZFunnel Visualization

Applies gsm.core::Analyze_NormalApprox_PredictBounds() to each month of dfAnalyzed, producing funnel plot bounds for each month.

Usage

PredictBounds_TimeZFunnel(
  dfAnalyzed,
  vThreshold = c(-3, -2, 2, 3),
  nMinSiteFraction = 0.2
)

Arguments

dfAnalyzed

A data frame output from Analyze_TimeZFunnel(). Must contain columns: NMonth, GroupID, GroupLevel, Numerator, Denominator, and Metric.

vThreshold

Numeric vector of threshold values for bounds. Default is c(-3, -2, 2, 3). A threshold of 0 (for the mean line) is automatically included by gsm.core::Analyze_NormalApprox_PredictBounds.

nMinSiteFraction

Minimum fraction of peak site count required to return non-NA bounds for a month. Months where the number of distinct GroupIDs is below this fraction of the maximum are returned with Metric = NA. Default is 0.2 (20%).

Value

A data frame with columns:

• NMonth: Time point (month number).

• Threshold: Threshold value (e.g., -3, -2, 0, 2, 3).

• Denominator: Sample size / exposure value.

• LogDenominator: Log of Denominator.

• Numerator: Predicted numerator at this threshold.

• Metric: Predicted Metric value at this threshold.

Multiple rows are returned per month per threshold, covering the range of Denominator values observed in the data.

See also

Analyze_TimeZFunnel() for calculating funnel scores, gsm.core::Analyze_NormalApprox_PredictBounds for the underlying bounds calculation.