Logical Invest at the Silicon Valley chapter of the investors AAII

What an audience and what an experience! Thanks to the investors AAII Silicon Valley!

As announced some weeks ago, on April 11 we hosted our first conference at the Silicon Valley chapter of the American Association of Individual Investors AAII in San Jose, CA.

Sharing and discussing some of the topics which are close to our heart with an incredible audience in live mode was such a great opportunity we completely overran the anticipated schedule by more than an hour. Deep-diving into special interests during follow up face-to-face meetings with individual investors and getting challenged by the computerized investment group helped to clarify expectations and ideas for further developments.

Recorded investors AAII conference

1.) AAII Board Introduction

Opening remarks by Lynn Gillette, President of the AAII SV, and Al Zmyslowski, head of the computerized investment group.

2) Who we are & What we stand for

Introduction of the Logical Invest team, what drives us, and some of our core beliefs in investing

3) Constructing your ‘all weather’ self-managed portfolio

The Logical Invest way of constructing portfolios using smart building blocks of assets, diversifiers, and different tactical approaches at the strategy and portfolio level

4) Building a well-balanced crash protection

Why Buy & Hold is dead and how to set up your portfolio for success even in the next correction

5) Harvesting the ‘Fear Premium’ and ‘Rebalancing losses’

Highly geeky ways to further improve your portfolio returns by investing in inverse volatility and shorting 3 times leveraged bear treasury ETF.

6) Markowitz Meets Logical Invest

How our own approach to portfolio construction evolved over time, and how to build dynamic weighted Meta-Strategies (aka Strategy of Strategies)

Or see the full investors AAII Youtube playlist list online.


8 thoughts on “Logical Invest at the Silicon Valley chapter of the investors AAII”

  1. Alex,

    I watched the videos of the conference this afternoon and thought you did a wonderful job! Obviously, you have a very technical background and you were able to put forth a presentation that was very coherent for the lay public.

    I wanted to ask you to amplify on one of the comments you made early on in the videos, regarding momentum. If I understood you correctly, you indicated that you felt momentum was important and that you always traded the top two performing strategies in 60/40 ratio.

    Could I ask you to elaborate further on exactly how you do this?

    Do you rebalance and re-assign weightings once monthly, I assume?
    Do all strategies have equal standing in the selection process, including those that have different iterations? (ie, BUG and BUG-LEV, GMR and GMRe, GSR and GSRlv, UIS and UIS-3x) — or are you making your selections from a more limited basket?
    Do you base selection just on last one month of performance, or quarterly, annually, etc?

    I’m very intrigued by this concept and was glad to see you were familiar with Gary Antonacci’s book “Dual Momentum Investing” as well as Meb Faber’s research. Clearly, you seem to have adopted the concept of cross-sectional momentum in your selection process. I’m assuming the concept of “longitudinal momentum” is less important here since you have a rotational model that is foundationally limiting your exposure to underperforming asset classes…

    Great work and can’t thank you guys enough for sharing and elaborating your thought processes.

    Kind regards,

    • Jeff,

      the ‘meta-strategy’ approach offers virtually unlimited possibilities of blending strategies together using a dynamic weighting, with many ‘knobs’ like different number and style of strategies, volatility scaling, min/max weighting limits to avoid corner solutions, sensitivity to changing volatility and market regimes…. the 60/40 momentum approach is just one example.

      We’re going to publish some meta-strategies with different approaches and different level of sophistication and aggresiveness initially, and then very much look forward to the voices and feedback from you on further variations to be published.

      Yes, we’re big fans of Meb and Gary, and think it’s just sound to grab some of the great ideas out there, instead of reinventing the wheel completely ourselves. To use similar language, ‘absolute momentum’ for us is when the return of the risky asset falls below our treasury diversifier (instead of the cash threshold), and relative momentum is intrinsically what we apply in the ‘modified Sharpe’ adaptive algorithm when we compare all possible pairs of risky assets and diversifiers.

      Hope this clarifies,

  2. Alex, enjoyed watching the videos. A couple of questions:

    I understand that several strategies move to cash in crash / high volatility circumstances. You mentioned in the talk that historic correlations (or inverse correlations) between various ETF’s may not be permanent going forward into the future. In the case of the paired switching/hedged strategies (Max Yield, UIS, etc.), what is the metric by which you monitor whether the inverse correlation exploited for each strategy is robust or weakening? What would be the circumstances under which you might replace an instrument within a strategy, or even to shut a strategy down due to a breakdown in the original premise of the strategy — for instance if both bonds and equities fall — not just for a day or two, but for a more prolonged period?

    The answer to the above leads to my next question which is if it would be possible to somehow communicate that metric to subscribers. Sort of a monthly robustness or correlation confidence monitor for each of the strategies when the signals are released. This was touched on recently if I recall when both ZIV and EDV fell on the same day and an email or blog post was sent out discussing this. One concern following the strategies is what happens if traditional relationships of different equity classes change — this was part of the explanation you gave why ultra-long back tests may not be appropriate in all cases. This might be another consideration when assembling a meta-strategy, not just investing in the two or three highest performing strategies, but in the strategies that are currently best meeting their stated objectives.

    My third question refers to the comments on the video regarding trading illiquid ETF’s and the possibility that the LogicalInvest trade signals could be moving the markets (maybe explains why ZIV dropped on April 1 ;). Anyway, is it not possible for you to have the signals computed on a daily basis and each subscriber could request the day of the month they chose to make their adjustment? Each subscriber would be sent a single monthly (or bimonthly for Max Yield) signal, but not necessarily on the same day. For your internal purposes you could report the CAGR statistics etc. for a 1st of the month rebalance, but the criticism of the commenter in the video would be addressed.

    Thanks for posting the videos,

    • David, thanks for the extensive comment!

      My comment was rather meant on the longer term ‘mechanism or premises’ our strategies are build on. Example, if suddenly the US or any other developed nation would go back to the Gold Standard, then probably our our ‘long term bond’ based crash-mechanism would not work anymore. Or other example, if the volatility futures would go into a backwardation mode , then our inverse volatility mechanism in the MYRS would stop working. So these are rather drastical events, which we cannot catch with pure math or hard thresholds or metrics, so we would communicate rather verbally to our followers – this is the ‘art’ of this business, beside all our quant approach. If in the short-term bonds correlate positively with equities over a couple of days like in Jan 2015, or over a couple of weeks like in May 2013 following FED statements , then this is just ‘normal course of nature’ – but no reason to abandon the underlying mechanism.

      As you mention, I’d love to come up with these ‘leading indicators’ when a strategy is expected to perform or not, but have not found such so far. So the best robustness indicator is the performance itself, as example when using the Top2 best performing out of 3 strategies. Recall that all our strategies are designed for a long-term investing process, with typical lookbacks of 60-80 days. As we’ve shown before, in our backtests any intent to react to shorter term events leads invariably to whipsaw effects. Using different tactical layers and diversification of the underlying assets and mechanism is the best protection from sharp corrections, and best for long term results.

      As example, here the rationale of how to use different tactical elements when building a portfolio
      – Pair switching of risky assets with bonds or inverse volatility as crash protection at the strategy level
      – Diversification by blending strategies with different asset classes / mechanism into portfolio (example Bond Rotation + UIS or MYRS)
      – At the portfolio level use momentum / max sharpe to shift allocation to ‘better performers’ or use Top x of y approach
      – Add volatility constraint to reduce exposure / investment when markets are in turmoil.

      Your idea of ‘spreading’ the execution over a longer time frame to avoid liquidity / spread or even availability concerns is interesting. I think part of that is already happening today, e.g. subscribers executing over a two days period after receiving our signals. But let me take up your idea for discussion, and will come back to you.

      Hope this clarifies,

  3. Hello Alex,
    I viewed yor presentation once again, and high marks for sharing and explaining some pretty powerful techniques. I am now a full subscriber and would like to review the slide deck from the AAII presentation rather than the video capture. Can you provide the link, please? I was unable to find it on the web site.

    Also, I remain interested in the meta-strategies and would like to be updated on them.

    Thanks, Richard

Leave a Comment