# Announcing the Portfolio Optimization Service

In 1952, a gentleman named Harry Markowitz put forth a theory exploring how one would apply mathematical theory to the management of investment securities, setting the stage for what is now referred to as Modern Portfolio Theory. In short, Harry posited that each individual investment – be it a stock, a bond, or a derivative – has a certain amount of risk, measured by its variance or how much its value moves around, and an expected return given that risk. The real insight, however, was to look at how a set of investments behaved collectively, instead of in isolation. This spurred the notion of diversification – you can smooth out your returns by holding things that move in opposition to each other, by holding assets that have negative correlations in industry parlance.

This notion is fairly intuitive. If you put all your eggs in one, very risky basket, you could make a lot of money if things turn out your way. You could lose tremendously if circumstances move against you. You can imagine it being a safer bet to proverbially spread your chips over the table instead of betting the farm on a single outcome. The crux of the theory is that there exists a diversified portfolio with an optimal expected return for an investor’s given level of risk. You just have to find it.

Enter the mathematicians. Seeking the right balance of investments results in a lot of choices. This might be further complicated by investor preferences, or investment theme affinity; not only do you have to find the optimal risk/return profile, you may have to do so while avoiding sin stocks, or focusing only on socially responsible investments. This suite of word problems has the same form as the traveling salesman problem and is solved the same way.

## Release of Portfolio Optimization Service

IBM has a rich history of solving exactly these kinds of problems. The class of problems, known as linear programming  are solved using optimization techniques. The CPLEX optimizer was pioneered in 1988 (and acquired by IBM in 2009) to solve these types of math problems. This same technology provides the technical foundation for the Portfolio Optimization service we’re releasing today, albeit applied in a financial context to aid our users in finding the best linear combinations of assets for their investment needs.

As with all of our financial APIs, we’re simplifying the process to democratize access to these complex and powerful investment tools. Never before has the investment management industry been able to provide this sophisticated functionality to users on the front lines who deal directly with customers. This type of functionality, coupled with the power of our other groundbreaking APIs, is poised to raise the bar of quality of analysis for the industry at large.

The first release of the Portfolio Optimization service will provide users the ability to construct investment portfolios based on objectives and constraints that matches the properties of a benchmark while taking into account a user’s custom preferences.

Consider the following use case:

Having just completed a survey to determine your investment goals and risk tolerance, your advisor – be they physical or virtual – offers a recommended set of investments for you. As you peruse the contents, you mention that you just can’t stomach the idea of giving your hard-earned money to companies who sell tobacco, given your family history. The advisor could remove such securities from the portfolio, but doing so may alter the properties of the collection in unanticipated ways – perhaps the tobacco stocks provided a diversification benefit, or brought down the average risk. Instead by providing the Portfolio Optimization service with the originally suggested portfolio, a list of eligible investments, and the user’s custom constraints, the service can return a portfolio that most closely resembles the original portfolio while still respecting the user’s preferences – all with a single API call!

The Portfolio Optimization service is available today on Bluemix. You can find it in the Experimental section, in the Finance subsection. It is an experimental service, so it might change over the coming months. For more information, see Getting Started with Portfolio Optimization.

Offering Manager, IBM Financial Risk APIs

Previous Post

Next Post

More stories
April 30, 2019

### Introducing IBM Analytics Engine v1.2 and Announcing the Deprecation of IBM Analytics Engine v1.0

We are excited to inform you about the new version of IBM Analytics Engine v1.2 that will be available starting May 15, 2019. Along with this release, Analytics Engine v1.0 will be retired.

April 23, 2019

### Announcing the Deprecation of the Watson Machine Learning JSON Token Authentication Service

We’d like to inform you about the deprecation of the Watson Machine Learning JSON Token Authentication service. This method of authentication will be retired on May 30, 2019.