Quantitative investing begins with an idea of how markets should react in various market conditions. It is understood that there is no single solution or Holy Grail for investing. The financial markets are too complex to be so easily conquered. Instead, one must seek to create innovative approaches to take advantage of the various identifiable market inefficiencies—trend following, factor analysis favoring small and value oriented investing, behavioral biases of investors that can be viewed as opportunities for profit, for example.
Once an idea is formed, a methodology is created to take advantage of it in the markets. But before that methodology can be implemented, it must be tested.
Flexible Plan Investments, Ltd. follows a rigorous testing protocol for each of its strategies. It begins with back-testing the methodology. Back-testing is essential to gain confidence in a trading approach and to allow for the development and introduction of new strategies. This back-testing process is done in two important ways.
WALK FORWARD ANALYSIS
This is generally recognized as the most unbiased form of back-testing. Rather than optimizing signals based upon the entire available history, the data is broken into shorter time periods. The best signals are selected for the earliest data in a sample’s time period. Then, using only the data available at the end of that historical time period, testing begins. The signals derived from the earlier period are walked forward (applied) through an entirely different time period and the results of using the signals on the investments being studied are recorded.
Then the process is repeated year-by-year or quarter-by-quarter or sometimes even week-by-week or month-by-month through the data history being examined—always applying the best signals from prior history to an entirely new set of data. This way you only apply the knowledge known before the test period occurred to the future data. The result is a back-tested researched return history free of the biases present in many over-optimized strategies.
TESTING FOR ROBUSTNESS
Even with walk forward back-testing, bias can still sneak into the process. One can use historical knowledge to in effect “stack the deck” by only applying the methodology to a universe of asset classes or funds that have a known successful history. Or one can repeatedly run the walk forward testing changing the methodology each time to better suit the data—eventually over fitting the data so that the resulting methodology has little application to a new untried environment.
To avoid these biases, Flexible Plan tests any resulting methodology on universes of investments that were not used, even in the walk forward testing, for the development of strategies. So, if a methodology works well with a universe of global investments, it must also prove effective applied unmodified to an investment portfolio made up of fixed income securities. By applying this multi-tiered testing approach, Flexible Plan gains insights into the likelihood of success of the methodology in future market environments.
NEW ADVANCES IN METHODOLOGY
This same methodology was applied to the development of FPI’s Evolution and Fusion strategies. In addition, new mathematical algorithms were created to further enhance the strategies’ robustness. These new algorithms employ a number of innovative features toward this end:
- A generalization process that forces the investment choices to seek a general rather than a specific solution
- Required usage of rank ordering of investment choices instead of simply processing raw numbers
- Usage of multiple, standard, non-optimized time windows for measurement purposes
- Utilizing return, correlation, and volatility to determine investment choices as well as position sizing
- Making investment choices based on the entire portfolio’s posture rather than solely on the evaluation of the individual asset class
The use of these algorithms is designed to make the resulting methodology more robust—anti-fragile, able to not only survive in new market environments, but thrive upon them.
FLEXIBLE PLAN STRATEGY INDICES
Flexible Plan makes available certain strategy indices, through an affiliate of the New York Stock Exchange, NYSE ARCA. To create these indices NYSE ARCA simply takes each strategy’s mathematical formula and applies it to actual investment data over a historical time period. It then replicates the same process each day and distributes the result. No attempt to optimize the results is made—it’s just the application of math to the data.
To create these indices, NYSE ARCA simply calculates each index’s value based on the index’s end-of-day holdings, as generated by the index’s mathematical rules. NYSE ARCA repeats the same process each day and distributes the result. No attempt is made to optimize the results.
These indices cannot be traded because indices are not tradable securities. Like research reports, the results are not the real result of actual trading. They are hypothetical because they are merely the application of modern methodology to historical data.
The FPI FusionSM , Self-adjusting Trend Following and Market Leaders Ultra Sectors Indexes (the “Indexes”) are calculated by NYSE Group, Inc. or its affiliates (“NYSE Group, Inc.”). The FPI FusionSM, Self-adjusting Trend Following and Market Leaders Ultra Sectors Index strategies or managed accounts, which are based on the Indexes, are not issued, sponsored, endorsed, sold or promoted by NYSE Group, Inc., and NYSE Group, Inc. makes no representation regarding the advisability of investing in such product.
NYSE GROUP, INC. MAKES NO EXPRESS OR IMPLIED WARRANTIES, AND HEREBY EXPRESSLY DISCLAIMS ALL WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE WITH RESPECT TO THE FPI FUSIONSM, SELF-ADJUSTING TREND FOLLOWING AND MARKET LEADERS ULTRA SECTORS INDEXES OR ANY DATA INCLUDED THEREIN. IN NO EVENT SHALL NYSE GROUP, INC. HAVE ANY LIABILITY FOR ANY SPECIAL, PUNITIVE, INDIRECT, OR CONSEQUENTIAL DAMAGES (INCLUDING LOST PROFITS), EVEN IF NOTIFIED OF THE POSSIBILITY OF SUCH DAMAGES.
Indexes do not predict or project the performance of an investment or investment strategy as one cannot invest directly in an index. While FPI’s Fusion strategies are based on the Fusion Indexes, FPI’s Self-adjusting Trend Following (“STF”) strategy is based on the STF Index and FPI’s Market Leaders Ultra Sectors strategy is based on the Market Leaders Ultra Sectors Index, the Indexes reflect the theoretical performance an investor would have obtained had it invested in the manner shown and does not represent returns an investor actually attained, as investors cannot invest directly in an index. No representation is being made that any client will or is likely to achieve results similar to those presented.
The Indexes’ calculations from January 2014 forward were performed by the NYSE Group, Inc.
From January 1998 to December 2013, the Indexes’ performance was calculated by FPI. The Indexes were constructed illustrating the hypothetical allocated performance of the Fusion methodology. The hypothetical performance of these allocations was drawn from various historical sources, including actual returns of FPI’s model account strategies, ETFs and open-end mutual funds (including long/short and/or leveraged funds).
From January 1998 to December 2013, there have been no assets managed under the Fusion Index rules at FPI. The Fusion mathematical algorithm has been in use by FPI since February 2013. The Self-adjusting Trend Following Index rules have been used to manage assets at FPI since July 2009. The Market Leaders Ultra Sectors Index rules have been used to manage assets at FPI since February 2014. FPI reserves the right to make enhancements to the Indexes’ methodologies.
The Indexes contain no management or advisory fees, other than the internal fees and expenses reflected in the NAV of the mutual funds or ETFs used. A client account in an FPI-offered Fusion, STF or Market Leaders Ultra Sectors strategy will incur advisory fees; additional fees may apply including transactions and trading costs determined by the custodian of the account. These fees and costs will decrease the return experienced by a client. Individual client account results will vary from the Indexes’ returns. Current and prospective clients should not assume that future performance will be the same or profitable. When provided, dividends are reinvested for indexes. In those cases where indexes do not provide dividend information, those returns would be understated. As individual tax rates vary, taxes have not been considered.
What historical performance results are available?
In addition to the individual account information provided to each client, FPI makes available two different sources of performance data.
First, there are GIPS® Composite Performance Reports. FPI claims compliance with the Global Investment Performance Standards (GIPS®) and has prepared and presented these reports in compliance with the GIPS standards. FPI has been is independently verified for the periods 1/1/08-12/31/13. A copy of the verification report(s) is/are available upon request. Verification assesses whether (1) the firm has complied with all the composite construction requirements of the GIPS standards on a firm-wide basis and (2) the firm’s policies and procedures are designed to calculate and present performance in compliance with the GIPS standards. Verification does not ensure the accuracy of any specific composite presentation.
Flexible Plan Investments, Ltd. is a federally registered independent investment adviser. Flexible Plan claims compliance with the Global Investment Performance Standards (GIPS). To receive a list of composite descriptions of Flexible Plan strategies and/or a compliant presentation, contact email@example.com or call 800-347-3539.
Second, Model Performance reports of actual accounts trading each of the strategies are available in two formats: a) they can be ordered for a specified time period or b) they are compiled in a mutual fund fact sheet manner where the results are provided for certain set time periods—year-to-date, last quarter, last 12 months, etc. Because Flexible Plan maintains scores of strategies on a wide range of investment platforms, requests for these reports are made on an individual basis rather than in an aggregate.
In choosing an asset management firm, it is essential to evaluate their research capabilities. Creating an investment strategy is not like producing a breakfast cereal, where you put the same flakes in the same box and distribute it every day.
Financial markets are diverse and ever changing. What works in one time period with one set of assets may not work in another. The strategies must be able to develop and change. That can only happen in a firm with a substantial research capability that can continually create, review, and seek to improve its strategies. Asset management firms should not believe that they can simply stand pat in the belief that they have the silver bullet to investment success. Instead, they must be flexible.