tech.nmfin.algoquant.demo.model.portfoliooptimization

Class RandomReturns

java.lang.Object

tech.nmfin.algoquant.demo.model.portfoliooptimization.RandomReturns

All Implemented Interfaces:

dev.nm.stat.random.Seedable

public class RandomReturns
extends Object
implements dev.nm.stat.random.Seedable

Generates a random return matrix.

Using this example. Generates an observations-by-assets matrix by multiplying random vectors by a Cholesky factor. NM matrix and vector indexing starts at one. Supply two seeds to seed(long...): the current implementation accesses seeds[1] even after its single-seed branch; this limitation is recorded in the audit notes.

Model and publication. Related theory and the scope of this example are distinguished below.

Implementation in this tutorial. This utility supplies artificial return observations rather than an economic model from a specific publication. A Cholesky factor introduces cross-asset dependence. The marginal generator and sample length remain explicit inputs, and the resulting matrix is not a historical performance series.

Author:

Haksun Li

Constructor Summary

Constructors

Constructor and Description
RandomReturns()
RandomReturns(dev.nm.stat.random.rng.univariate.RandomNumberGenerator rng1)

Method Summary

Modifier and Type	Method and Description
dev.nm.algebra.linear.matrix.doubles.Matrix	getRandomCovariance(int N)
dev.nm.algebra.linear.matrix.doubles.Matrix	getRandomReturns(int N, int T)
dev.nm.algebra.linear.matrix.doubles.Matrix	getRandomReturns(dev.nm.algebra.linear.matrix.doubles.Matrix S, int T) Applies a Cholesky factor of S to generated vectors.
void	seed(long... seeds) Seeds the return and correlation generators.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

RandomReturns

public RandomReturns(dev.nm.stat.random.rng.univariate.RandomNumberGenerator rng1)

RandomReturns

public RandomReturns()

Method Detail

getRandomCovariance

public dev.nm.algebra.linear.matrix.doubles.Matrix getRandomCovariance(int N)

getRandomReturns

public dev.nm.algebra.linear.matrix.doubles.Matrix getRandomReturns(dev.nm.algebra.linear.matrix.doubles.Matrix S,
                                                                    int T)

Applies a Cholesky factor of S to generated vectors.

Parameters:

S - positive-definite square covariance input (one row/column per asset)

T - number of generated observations

Returns:

a T-by-S.nRows() matrix; NM indices start at 1

getRandomReturns

public dev.nm.algebra.linear.matrix.doubles.Matrix getRandomReturns(int N,
                                                                    int T)

seed

public final void seed(long... seeds)

Seeds the return and correlation generators.

Specified by:

seed in interface dev.nm.stat.random.Seedable

Parameters:

seeds - two seeds, first for rng1 and second for rng2; the current implementation does not safely support fewer than two seeds