classification: Classify Effects by Comparing the Confidence Intervals with a...

In inbo/effectclass: Classification and Visualisation of Effects

Classify Effects by Comparing the Confidence Intervals with a Reference and Thresholds

Description

• ⁠++⁠ strong positive effect: max(threshold) < lcl

• + positive effect: ⁠reference < lcl < max(threshold)⁠ and max(threshold) < ucl

• ⁠+~⁠ moderate positive effect: reference < lcl and ucl < max(threshold)

• ~ no effect: ⁠min(threshold) < lcl < reference⁠ and ⁠reference < ucl < max(threshold)⁠

• ⁠-~⁠ moderate negative effect: min(threshold) < lcl and ucl < reference

• - negative effect: lcl < min(threshold) and ⁠min(threshold) < ucl < reference⁠

• ⁠--⁠ strong negative effect: ucl < min(threshold)

• ⁠?+⁠ potential positive effect: ⁠min(threshold) < lcl < reference⁠ and max(threshold) < ucl

• ⁠?-⁠ potential negative effect: lcl < min(threshold) and ⁠reference < ucl < max(threshold)⁠

• ⁠?⁠ unknown effect: lcl < min(threshold) and max(threshold) < ucl

Usage

classification(lcl, ucl, threshold, reference = 0)

Arguments

lcl	A vector of lower confidence limits.
ucl	A vector of upper confidence limits.
threshold	A vector of either 1 or 2 thresholds. A single threshold will be transformed into reference + c(-abs(threshold), abs(threshold)).
reference	The null hypothesis. Defaults to 0.

See Also

Other classification functions: coarse_classification(), remove_sign()

inbo/effectclass documentation built on June 8, 2024, 2:43 a.m.