- The new PSS Software Package is the most powerful investment diversification and asset allocation software package available today. PSS gives the user more quantitative portfolio selection ability than is currently used by the largest and most sophisticated investment funds.
- PSS is portfolio optimization and construction software that uses the Digital Portfolio Theory model to add an analytic measure of memory to Modern Portfolio Theory. Portfolio return is maximized while constraining portfolio mean-reversion variance, covariance, and autocovariance in a stand-alone desktop application.
- The PSS software package offers substantial capability for investment managers and pension sponsors to analyze and optimize portfolios of assets using the latest digital signal processing technology.
- The PSS Software Package utilizes a completely new form of stochastic programming particularly and uniquely suited to financial random process. The Digital Portfolio Theory (DPT) signal-processing model of optimization under uncertainty has not previously been applied to financial systems.
- Long term financial risk is broken into monthly, quarterly, 6-month, annual, 2-year and 4-year mean-reversion variance components. DPT adjusts risk exposure from calendar length periods to maximize return and reduce risk over the investors holding period.
- The PSS application simultaneously performs optimal fundamental analysis using earnings growth, change in earnings growth, dividend yield, P/E ratios, market capitalization or firm size, and book to market value.
- Digital Portfolio Theory promises to advance portfolio management to the same extent that digital systems theory is advancing speech processing, image processing, and other applications.
- Large investment management firms, institutional managers, financial advisors, and third party software vendors will be interested in integrating the DPT optimization model into their existing research and applications.
PSS supplies state-of-the-art investment selection technology by utilizing the new DPT in conjunction with the latest digital signal processing technology to create optimal and reliable portfolio performance given the objectives and holding period of clients. PSS Rel. 2.0 Digital Portfolio Theory software is the only digital signal processing application specifically designed for financial return processes.
PSS Software Requirements:
- requires PC platform
- requires Windows
- requires Spreadsheet (Excel or Lotus 1-2-3) for Windows
- requires What’sBest! Spreadsheet Solver Add-in by Lindo Systems
- stand-alone, fully contained portfolio optimization system that does not require internet connection
PSS Software Characteristics:
- solves the return maximization investment optimization problem for target levels of mean-reversion risk
- can select portfolios from an 8000 alternative security universe
- efficient portfolios are found using non-parametric linear programming
- more stable and reliable portfolios are found than using MPT since quadratic programming is not used
- mean-reversion variances, and cross correlations are estimated from historical data using digital signal processing
- measures mean-reversion correlation and autocorrelation risk relative to the index the user specifies as a performance benchmark
- controls systematic and unsystematic mean-reversion risk, and calendar and non-calendar mean-reversion risk of efficient portfolios
- finds optimal portfolios based on earnings growth, change in earning growth, dividend yield, P/E ratios, size, and book to market value
- controls solution portfolio CAPM calendar length mean-reversion betas
PSS Portfolio Selection System Release 2.0 is a spreadsheet software package for Windows that utilizes the Digital Portfolio Theory. PSS finds optimal portfolios by maximizing portfolio return while constraining calendar, non-calendar, systematic and unsystematic mean-reversion risks. Fundamental variables are simultaneously constrained and the user can add additional constraints. The PSS User’s Guide provides the mathematical definition of the Digital Portfolio Theory model as well as valuable intuition about efficient market timing.
The PSS software package is supported by the textbook PORTFOLIO MANAGEMENT.
- PSS uses Digital Signal Processing (DSP) to measure mean-reversion variance spectral density and cross correlations of asset returns to find stable calendar and non-calendar mean-reversion risk estimates.
- PSS uses 16 years of monthly time series security returns or prices.
- PSS uses a 48-month return signal length to capture long-term mean-reversion risk to accommodate long holding periods.
- PSS uses the Welch DSP method to achieve highly significant calendar mean-reversion variance estimates.
- PSS uses a rectangular window for narrow spectral resolution.
- Welch DSP samples multiple realizations of return processes resulting in highly consistent and stable estimates for more confident input data to the optimization process. Other methods do not include the memory of calendar and non-calendar mean-reversion risks.
- Linear programming optimization used in DPT results in more robust diversified portfolios that depend on investor holding period.
User Supplied Data:
- PSS is compatible with any historical security database but does not supply financial data.
- The user supplies a spreadsheet file to PSS for each alternative security (stock, ETF, mutual fund, asset class, index, etc.).
- Spreadsheet files must contain 16 years of monthly time series security returns, or prices.
- The user must specify the index that will be used as the benchmark to estimate systematic risk.
- Each spreadsheet file may contain fundamental variable information for inclusion in the optimization.
- Historical means and variances can be scaled or modified to reflect individual forecasts before optimization.
- PSS provides an interface to Standard & Poor’s Compustat database.
Capabilities and Characteristics of the Digital Portfolio Theory Solutions:
Portfolio Construction and Programming:
- solves for efficient long-term investment portfolios
- finds efficient long-short position portfolio solutions
- finds efficient arbitrage portfolios
- controls calendar length mean-reversion risk
- controls non-calendar length mean-reversion risk
- controls long memory risk up to 8 years
Portfolio Selection Strategy Support:
- supports single period, mean-variance (MV) portfolio selection
- useful for efficient asset class allocation mix
- useful for efficient industry, or sector allocation
- useful for efficient style, or objective allocation
- can be used to maximize alpha and control tracking error
- PSS is appropriate for long-term investment decisions only, from 1 month to 8 years
- PSS computes long-term historical estimates of mean, mean-reversion variance, and autocorrelation.
- includes mean-reversion risk, market risk, and fundamental risk in the optimization decision
- PSS cannot be used as a day trading tool
- PSS is not a technical analysis tool
- PSS is not an options trading tool
Timing Risk Analysis:
- controls January effect risk
- controls annual risk effects
- controls quarterly risk effects
- controls monthly risk effects
- controls presidential effect risk
- controls systematic risk
- controls unsystematic risk
- controls CAPM betas
- controls calendar mean-reversion betas
- controls long memory mean-reversion betas
Fundamental Analysis – PSS includes constraints that simultaneously control:
- earnings growth
- change in earning growth
- market capitalization, or firm size
- PE ratio relative to industry PE ratio
- dividend price ratio
- book to market value ratio
- additional fundamental variables can be added
Portfolio Management Constraints – optionally added by the user:
- budget constraints
- trading cost constraints
- portfolio size constraints
- liquidity constraints
- turnover constraints
- lower and upper bound constraints
- integer variables can be added to control fixed costs
In addition to the PSS software package site license, Portfolio Selection Systems provides ongoing portfolio optimization support and advice on a contract basis.
Portfolio Selection System: Digital Portfolio Theory
ISBN 977-19-3745-6 First Edition (1997)
1. Portfolio management.
2. Investment analysis.
Copyright © C. Kenneth Jones, 2018
For information: firstname.lastname@example.org