Skip to content

stat

The stat package provides descriptive statistics functions for numeric lists: measures of central tendency, spread, distribution, correlation, and normalization.

Package Functions

Basic Measures

Function Returns Description
sum(data) integer\|float Sum of all values
min(data) integer\|float Minimum value
max(data) integer\|float Maximum value
range(data) integer\|float max - min
mean(data) float Arithmetic mean
median(data) integer\|float Middle value (or average of two middle values)
mode(data) list List of most-frequent values
sorted(data) list Sorted copy (does not mutate the original)

Spread

Function Returns Description
variance(data, population?) float Sample variance by default; pass true for population variance
stdev(data, population?) float Sample standard deviation; pass true for population
percentile(data, p) float p-th percentile in [0, 100] via linear interpolation
zscore(data) list List of z-scores (standard scores)

Bivariate

Function Returns Description
covariance(x_data, y_data) float Sample covariance of two lists
correlation(x_data, y_data) float Pearson correlation coefficient [-1, 1]

Distribution

Function Returns Description
frequency(data) hashmap Maps each unique value (as string) to its count
normalize(data) list Rescales values to [0.0, 1.0] range

Function Details

percentile(data, p)

Parameters

Type Name Description
list data List of numeric values
integer\|float p Percentile in [0, 100]

Returns float

variance(data, population?) / stdev(data, population?)

Parameters

Type Name Description Default
list data List of numeric values —
boolean population true for population statistic, false for sample false

covariance(x_data, y_data) / correlation(x_data, y_data)

Both lists must have the same length.

Examples

import "stat"

data = [4, 7, 13, 2, 1, 9, 7, 3]

println stat::sum(data)        # 46
println stat::min(data)        # 1
println stat::max(data)        # 13
println stat::range(data)      # 12
println stat::mean(data)       # 5.75
println stat::median(data)     # 5.5
println stat::mode(data)       # [7]

Spread

import "stat"

data = [2.0, 4.0, 4.0, 4.0, 5.0, 5.0, 7.0, 9.0]

println stat::variance(data)              # 4.571... (sample)
println stat::variance(data, true)        # 4.0      (population)
println stat::stdev(data)                 # 2.138...

println stat::percentile(data, 25)        # 3.5
println stat::percentile(data, 75)        # 5.5

scores = stat::zscore(data)
println scores   # [-1.16, -0.46, ...]

Correlation

import "stat"

x = [1, 2, 3, 4, 5]
y = [2, 4, 5, 4, 5]

println stat::covariance(x, y)    # 1.5
println stat::correlation(x, y)   # 0.874...

Frequency and normalization

import "stat"

data = [1, 2, 2, 3, 3, 3, 4]

freq = stat::frequency(data)
println freq    # { "1": 1, "2": 2, "3": 3, "4": 1 }

normalized = stat::normalize(data)
println normalized  # [0.0, 0.333..., 0.333..., 0.666..., 0.666..., 1.0, ...]