concentration_index_positive: Inclusion-fragility concentration index (positive weights...

In mispitools: Missing Person Identification Tools

Inclusion-fragility concentration index (positive weights only)

Description

Signed variant of concentration_index that uses only the supporting (w_k > 0) contributions. For forensic inclusion cases, this measures the fragility of the support for the prosecution hypothesis specifically, ignoring any excluding markers.

Usage

concentration_index_positive(weights)

Arguments

weights

Numeric vector of per-step contributions (typically \log_{10} \mathrm{LR}_k).

Details

Defined as

C_W^+(w) = \frac{\max_k w_k^+}{\sum_k w_k^+}

with w_k^+ = \max(w_k, 0). Under Theorem T-003.1 of the Belief Dynamics framework, C_W^+ equals the fraction of the total positive weight of evidence lost under adversarial removal of the single most impactful supporting marker.

Value

Numeric scalar in [0, 1]. Returns 0 if no weight is positive.

References

Marsico, F. L. & Egeland, T. (in preparation). Belief dynamics during the investigative process.

See Also

concentration_index, leave_one_out.

Examples

# Mostly supporting, one dominant
concentration_index_positive(c(2.5, 0.3, 0.4, 0.2, 0.1))

# Mixed with one strong exclusion: the exclusion does NOT contribute
concentration_index_positive(c(1.0, 0.8, 0.9, -5.0, 1.1))

mispitools documentation built on Aug. 26, 2026, 1:08 a.m.