tech.nmfin.algoquant.demo.model.portfoliooptimization

Class PortfolioOptimComparisons1Demo

java.lang.Object

tech.nmfin.algoquant.demo.model.portfoliooptimization.PortfolioOptimComparisons1Demo

@Demo
public class PortfolioOptimComparisons1Demo
extends Object

This demo shows how to build customized portfolio optimization algorithms and call these optimizers for the same data set. AQ has a suite of tools to do asset allocation/portfolio optimization. It comes with out-of-box ready to use tools and users can also put together different models to customize their own optimizers. This demo shows a few simple out-of-box optimizers. The simulated returns are from beta distribution and are correlated so the regular smart beta algorithms don't exhibit usual behaviors when uncorrelated normal distributions are assumed.

Using this example. Feeds the same generated return matrix to several optimizer implementations to compare their weights. Rows are observations and columns are assets; keep this order when wrapping a result in PortfolioWeights. This is a numerical comparison without a transaction-cost or execution simulation.

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

Implementation in this tutorial. Compares multiple optimizer families on one synthetic return matrix. Black–Litterman combines equilibrium information with views. The wrapper assumes equal market weights, sample means as views and tau=1/T; this demo supplies risk aversion rather than explicit investor views. Other branches use Markowitz, Corvalán, smart-beta assumptions and Lai-style resampling. These are separate models, not interchangeable solver settings.

Author:

Haksun Li

See Also:

F. Black and R. Litterman (1992), Global Portfolio Optimization, Financial Analysts Journal 48(5), 28–43, H. Markowitz (1952), Portfolio Selection, Journal of Finance 7(1), 77–91, A. Corvalán (2005), Well Diversified Efficient Portfolios, Central Bank of Chile Working Paper 336, T. L. Lai, H. Xing and Z. Chen (2011), Mean–variance portfolio optimization when means and covariances are unknown, Annals of Applied Statistics 5(2A), 798–823

Constructor Summary

Constructors

Constructor and Description
PortfolioOptimComparisons1Demo()

Method Summary

Modifier and Type	Method and Description
static tech.nmfin.portfoliooptimization.PortfolioOptimizationAlgorithm	getBlackLitterman(double riskAverse)
static tech.nmfin.portfoliooptimization.PortfolioOptimizationAlgorithm	getCorvalan2005(double riskAverse, double relaxation)
static tech.nmfin.portfoliooptimization.PortfolioOptimizationAlgorithm	getCorvalanSAAM(double relaxation)
static tech.nmfin.portfoliooptimization.PortfolioOptimizationAlgorithm	getEqualCor(double riskAverse)
static tech.nmfin.portfoliooptimization.PortfolioOptimizationAlgorithm	getEqualMean(double riskAverse)
static tech.nmfin.portfoliooptimization.PortfolioOptimizationAlgorithm	getEqualWeights(double riskAverse)
static tech.nmfin.portfoliooptimization.PortfolioOptimizationAlgorithm	getMarkowitz(double riskAverse)
static tech.nmfin.portfoliooptimization.PortfolioOptimizationAlgorithm	getSAAM()
static tech.nmfin.portfoliooptimization.PortfolioOptimizationAlgorithm	getSOCP(double riskAverse)
static void	main(String[] args)
static dev.nm.algebra.linear.matrix.doubles.Matrix	randomReturnMatrix(int N, int T) Generates a random return matrix.

Methods inherited from class java.lang.Object

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

Constructor Detail

PortfolioOptimComparisons1Demo

public PortfolioOptimComparisons1Demo()

Method Detail

main

public static void main(String[] args)
                 throws Exception

Throws:

Exception

getMarkowitz

public static tech.nmfin.portfoliooptimization.PortfolioOptimizationAlgorithm getMarkowitz(double riskAverse)
                                                                                    throws Exception

Throws:

Exception

getCorvalan2005

public static tech.nmfin.portfoliooptimization.PortfolioOptimizationAlgorithm getCorvalan2005(double riskAverse,
                                                                                              double relaxation)
                                                                                       throws Exception

Throws:

Exception

getSOCP

public static tech.nmfin.portfoliooptimization.PortfolioOptimizationAlgorithm getSOCP(double riskAverse)
                                                                               throws Exception

Throws:

Exception

getEqualCor

public static tech.nmfin.portfoliooptimization.PortfolioOptimizationAlgorithm getEqualCor(double riskAverse)
                                                                                   throws Exception

Throws:

Exception

getEqualMean

public static tech.nmfin.portfoliooptimization.PortfolioOptimizationAlgorithm getEqualMean(double riskAverse)
                                                                                    throws Exception

Throws:

Exception

getEqualWeights

public static tech.nmfin.portfoliooptimization.PortfolioOptimizationAlgorithm getEqualWeights(double riskAverse)
                                                                                       throws Exception

Throws:

Exception

getSAAM

public static tech.nmfin.portfoliooptimization.PortfolioOptimizationAlgorithm getSAAM()
                                                                               throws Exception

Throws:

Exception

getCorvalanSAAM

public static tech.nmfin.portfoliooptimization.PortfolioOptimizationAlgorithm getCorvalanSAAM(double relaxation)
                                                                                       throws Exception

Throws:

Exception

getBlackLitterman

public static tech.nmfin.portfoliooptimization.PortfolioOptimizationAlgorithm getBlackLitterman(double riskAverse)
                                                                                         throws Exception

Throws:

Exception

randomReturnMatrix

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

Generates a random return matrix.

Parameters:

N - the number of stocks

T - the number of observations

Returns:

a random return matrix