The Selection Advantage
Most Robo Advisors will put your account into a set number of ETF’s and never change your positions. This is a classic investment technique called “buy and hold.” As long as the market, and all of your holdings, continuously make gains, that is a fine approach. However, we all that this is not the case. Markets change, and it makes absolute sense that our positions should change with them.
OmniFunds is different. We evaluate positions at a determined interval (weekly, monthly, etc.). We use a filtering and ranking process at each interval in order to see which securities the fund should be in based on their current performance. This allows us to move into the positions that are currently poised to help us reach our investment goals.
The Omnifund Architecture
This white paper explains the security selection process. In order to understand this process, it is important to understand the structure of an Omnifund. Each Omnifund consists of a Portfolio. A Portfolio consists of any number of Sectors. Some Sectors may have one Sector, while others may have many more Sectors.
Sectors are independent of each other. Each Sector contains a unique universe of securities. These securities are analyzed at every interval, as defined in the Omnifund. This analysis includes a filtering and ranking process in order determine which securities will be used by the Omnifund for the current interval.
The Filtering Process
The first step in the selection process is to filter out securities that we do not want to be considered for our OmniFund. This is done order to better define the current market behavior of the securities we want in our OmniFund. Filters are applied to the universe of securities in a particular OmniFund sector.
Our filtering process is done by applying technical indicators to the securities in order to see if they meet the defined criteria. For example, we may want to only consider stocks or ETF’s that are currently in a long term upward trend. A common indicator used to determine long term trend is the 200 Period Simple Moving Average (SMA). If current price is above the 200 Period SMA, the security is consider to be in a long term upward trend. Conversely, if the current price is below the 200 Period SMA, the security is considered to be in a downward trend. So in order to only consider securities that are in a long term upward trend, we would only consider the stocks and ETF’s that are currently trading above their 200 Period SMA.
An Omnifund Sector is not required to use a filter, although most of them do. For more information on which Omnifund Sectors are using a filter and what filter they are using, please visit the OmniFunds details page.
Once we have filtered out the securities that do not meet our filtering criteria, OmniFunds will then rank the remaining securities in order to determine which ones will be held as positions over the next interval. All of the current OmniFunds use classic technical indicators in order rank securities.
Depending the investment goal of the Omnifund, we would rank the securities in different ways. For example, one Omnifund may want find securities that currently oversold from a short term perspective, so it might use a 9 Period Relative Strength Index (RSI) and rank securities in ascending order. Another Omnifund may want to find the securities with the strongest medium term momentum, so it would use a 40 Period Momentum indicator and rank in descending order.
Realize that for every Omnifund, there are multiple sectors. Each sector has its own ranking function. While the same ranking function can be used in multiple sectors, there will be times when different sectors in the same Omnifund are using different sectors. For more information on which ranking functions are used for a particular sector in an Omnifund, please visit the Omnifund details page.
Ranking Based on Probability
Of course, the positions that we really want to take are the ones that have a high probability of posting gains. Our OmniFunds developers are currently working on multiple Artificial Intelligence components, including Genetic Algorithms and Neural Networks. These components will look at the universe of securities in an Omnifund and rank them based on a “probability score”. This superior approach to determining which positions we should be holding gives OmniFunds a distinct advantage over the vast majority of Robo Advisors.