kannisto: Kannisto's Age Heaping Index

In heaping: Correction of Heaping on Individual Level

Kannisto's Age Heaping Index

Description

Kannisto's index compares the count at a specific age to a geometric mean of surrounding ages, providing a measure of heaping that is robust to exponentially declining populations at old ages.

Usage

kannisto(x, Agei = 90, weight = NULL)

Arguments

x	numeric vector of individual ages.
Agei	single age value to evaluate (default 90).
weight	optional numeric vector of sampling weights.

Details

Calculate Kannisto's index for detecting heaping at a specific old age.

Unlike other indices that use arithmetic means, Kannisto's index uses geometric means of neighboring ages, which is more appropriate for old-age populations where counts decline exponentially.

The index is calculated as the ratio of the count at age Agei to the geometric mean of counts at ages Agei-2 through Agei+2.

Interpretation:

• 1.0: no heaping at the specified age

• >1.0: heaping (attraction to the age)

• <1.0: avoidance of the age

Value

A single numeric value representing Kannisto's index.

Author(s)

Matthias Templ

References

Kannisto, V. (1999). Assessing the information on age at death of old persons in national vital statistics. Validation of Exceptional Longevity, Odense Monographs on Population Aging, 6, 235-249.

See Also

jdanov for Jdanov's index, coale_li for Coale-Li index.

Other heaping indices: bachi(), coale_li(), heaping_indices(), jdanov(), myers(), noumbissi(), spoorenberg(), whipple()

Examples

# Create old-age data with heaping at 90
set.seed(42)
age <- c(sample(85:95, 2000, replace = TRUE),
         rep(90, 200))  # Add heaping at 90
kannisto(age, Agei = 90)  # Should be > 1

# No heaping
age_uniform <- sample(85:95, 2000, replace = TRUE)
kannisto(age_uniform, Agei = 90)  # Should be close to 1

heaping documentation built on Feb. 10, 2026, 1:08 a.m.