APT Correlation Allocation Overview

Classic diversification techniques (e.g. MPT) will classify a group of securities into “groups” and allocate based on historical returns. This might sound good in theory, but in practice there is no guarantee that these groups are not highly correlated and that we are diversified.

The Advanced Portfolio Theory (APT) Correlation Allocation Method takes a two-step approach to diversification in order to achieve two goals. First, we want to be able to achieve true diversification regardless of a securities group classification. Secondly, we want to allocate our portfolio based on probability of future returns, not based on historical performance.

The first step in the process is to measure the correlation of each security in a portfolio to a symbol, or index, that represents the broad market. This is known as the Single-Index Model of diversification. Once we know how correlated our individual securities are to an index, we separate the securities into bins based on their individual correlation to the broad market. To do so, we look at the highest and lowest correlation coefficients of the securities, and then create a binned scale between the values. We then bin the securities according to where their correlation coefficients fall in the scale.

Now that our securities are binned, we need to determine which bins of securities have the highest probability of producing gains. We rank our securities based on a measurement. For basic OmniFunds, this could be the value of a simple indicator. For more advanced OmniFunds, we would use Artificial Intelligence to determine which bin of securities has the highest profit potential.

Once we know which correlation bins have the highest ranking, we then allocate to the securities in each bin based on the collective rank. So, the bin with the highest rank gets the highest allocation. The bin with the second highest rank gets the second highest allocation, and so on.

At this time, we are using 4 allocation bins with the following allocation values:

Bin Ranking

Portfolio Allocation Rercentage

1

40%

2

25%

3

20%

4

15%

The number of bins and the allocation per bin are currently fixed. Moving ahead, we are planning to make both of these settings configurable to the user.

Now that the allocation per bin is established, we then take the determined allocation for that bin and we allocate to each security in that bin according to the security’s rank. The individual rankings are used to distribute the bins total allocation so that the highest ranked security gets the highest allocation, the second ranked security gets the second most allocation, and so on. The actual allocation percentage per security is determined based on the security’s ranking relative to the total allocation for the bin.

The Correlation Allocation approach to portfolio diversification is unique to OmniFunds, and yet another advantage that OmniFunds has other automated investment services. 


This website is operated and maintained by Intelligent Fund Management LLC. Brokerage services provided to clients of Intelligent Fund Management LLC by Gar Wood Securities, an SEC registered broker-dealer and member FINRA/SIPC.Investments: Not FDIC Insured • No Bank Guarantee • May Lose Value. Investing in securities involves risks, and there is always the potential of losing money when you invest in securities. Before investing, consider your investment objectives and Omnifund’s charges and expenses. Past performance does not guarantee future results, and the likelihood of investment outcomes are hypothetical in nature. See full disclosures for more information. Not an offer, solicitation of an offer, or advice to buy or sell securities in jurisdictions where Omnifunds is not registered. Contact: Intelligent Fund Management, 9111 Jollyville Rd, Suite 275 Austin, TX 78759. Tel: 800-880-0338