Transition Analysis Age Estimation Calculator
Density score
Morphology stages

Method and applicability

Maximum composite-likelihood age from threshold-specific cumulative-probit transition models. Bilateral observations are combined by geometric mean within each trait.

Reference sample: 533 females and 605 males aged 45 or older (n = 1,138). The pooled model has no sex term and has not been validated for males younger than 45.

Maximum-likelihood age
—
years
90%
—
95%
—
Relative likelihood of age Relative likelihood curve with maximum-likelihood age and calibrated 90 and 95 percent interval bars.

Enter all six traits to calculate an estimate.

Intervals are individual age-estimation intervals calibrated to held-out coverage; they are not confidence intervals for a fitted regression coefficient.

Age Likelihood Ratio Calculator

Pooled transition reference: n = 1,138 — 533 all-age females + 605 males aged 45+

The evidential proposition is centered on the estimated mean age. H₁ is fixed as actual age within ±2 SD of the estimated mean.

Enter observations


Trait Left Right

At least one side is required for each bilateral trait. When both are entered, the two likelihoods are combined by geometric mean.

Age estimate and fixed ±2 SD proposition

Estimated (mean) age
—
Age-profile standard deviation
—
Fixed tolerance
—

H₁
—
H₂
—

Likelihood ratio

Composite proposition LR
—
log₁₀(LR)
—
Direction
—
Supports H₂ LR = 1 Supports H₁

Maximum-likelihood age
—
90% calibrated LR interval
—
95% calibrated LR interval
—

—
Interpretation

H₁ = actual age lies within two likelihood-profile standard deviations of the estimated mean age.

H₂ = actual age lies outside that interval but remains within the applicable reference age range.

This is a research-use composite likelihood ratio, not a validated forensic DNA LR. Report that the observed morphology is X times more probable under one proposition than the other. Do not state that an age itself is X times more likely to be correct.

Model details

Reference sample: 1,138 individuals — 533 females and 605 males aged 45 or older. The pooled model has no sex coefficient or male offset. Sex selection controls the applicable age range only.

Threshold-specific cumulative-probit transition curves are used for Density, UB, LB, Outline, Surface Texture, and Topography. Composite likelihood is evaluated on a 0.25-year age grid.