Skip to content
Ubiquev1.0.4
functionparamvar
paramvar(
mu: number,
sigma: number,
p?: number,
amount?: number,
period?: number
): number

Parametric Value-At-Risk.

Parametric Value-At-Risk assuming returns are normally distributed. It can work with numbers, arrays, row vectors, and column vectors.

Normal loss model: mean scales with period, volatility with sqrt(period). Positive results represent losses; negative results are possible. Shapes are preserved.

Examples

VaR with default parameters

import { assertEquals, assertAlmostEquals } from "jsr:@std/assert";

assertAlmostEquals(paramvar(0, 1), 1.6448536269514722, 8.224268134757361e-13);

VaR with arrays

import { assertEquals, assertAlmostEquals } from "jsr:@std/assert";

const expected = [1.6448536269514722, 3.2897072539029444, 4.934560880854417];
paramvar([0, 0, 0], [1, 2, 3]).forEach((v, i) => assertAlmostEquals(v, expected[i], 3e-12));

Parametric VaR for a single asset

import { assertEquals, assertAlmostEquals } from "jsr:@std/assert";

assertAlmostEquals(paramvar(0.0179, 0.023230487630602065), 0.020310751835047117, 1.0155375917523558e-14);

Parameters

mu: number

mean value (def: 0)

sigma: number

standard deviation (def: 1)

optional
p: number

VaR confidence level in range (0,1) (def: 0.95)

optional
amount: number

portfolio/asset amount (def: 1)

optional
period: number

time horizon (def: 1)

Return Type

number

Parametric Value-At-Risk

paramvar(
mu: array,
sigma: array,
p?: number,
amount?: number,
period?: number
): array

Parametric Value-At-Risk.

Parametric Value-At-Risk assuming returns are normally distributed. It can work with numbers, arrays, row vectors, and column vectors.

Normal loss model: mean scales with period, volatility with sqrt(period). Positive results represent losses; negative results are possible. Shapes are preserved.

Examples

VaR with default parameters

import { assertEquals, assertAlmostEquals } from "jsr:@std/assert";

assertAlmostEquals(paramvar(0, 1), 1.6448536269514722, 8.224268134757361e-13);

VaR with arrays

import { assertEquals, assertAlmostEquals } from "jsr:@std/assert";

const expected = [1.6448536269514722, 3.2897072539029444, 4.934560880854417];
paramvar([0, 0, 0], [1, 2, 3]).forEach((v, i) => assertAlmostEquals(v, expected[i], 3e-12));

Parametric VaR for a single asset

import { assertEquals, assertAlmostEquals } from "jsr:@std/assert";

assertAlmostEquals(paramvar(0.0179, 0.023230487630602065), 0.020310751835047117, 1.0155375917523558e-14);

Parameters

mean value (def: 0)

sigma: array

standard deviation (def: 1)

optional
p: number

VaR confidence level in range (0,1) (def: 0.95)

optional
amount: number

portfolio/asset amount (def: 1)

optional
period: number

time horizon (def: 1)

Return Type

Parametric Value-At-Risk

paramvar(
mu: matrix,
sigma: matrix,
p?: number,
amount?: number,
period?: number
): array | matrix

Parametric Value-At-Risk.

Parametric Value-At-Risk assuming returns are normally distributed. It can work with numbers, arrays, row vectors, and column vectors.

Normal loss model: mean scales with period, volatility with sqrt(period). Positive results represent losses; negative results are possible. Shapes are preserved.

Examples

VaR with default parameters

import { assertEquals, assertAlmostEquals } from "jsr:@std/assert";

assertAlmostEquals(paramvar(0, 1), 1.6448536269514722, 8.224268134757361e-13);

VaR with arrays

import { assertEquals, assertAlmostEquals } from "jsr:@std/assert";

const expected = [1.6448536269514722, 3.2897072539029444, 4.934560880854417];
paramvar([0, 0, 0], [1, 2, 3]).forEach((v, i) => assertAlmostEquals(v, expected[i], 3e-12));

Parametric VaR for a single asset

import { assertEquals, assertAlmostEquals } from "jsr:@std/assert";

assertAlmostEquals(paramvar(0.0179, 0.023230487630602065), 0.020310751835047117, 1.0155375917523558e-14);

Parameters

mean value (def: 0)

sigma: matrix

standard deviation (def: 1)

optional
p: number

VaR confidence level in range (0,1) (def: 0.95)

optional
amount: number

portfolio/asset amount (def: 1)

optional
period: number

time horizon (def: 1)

Return Type

Parametric Value-At-Risk