|
About IndexInvestor.com |
Privacy Policy |
Transaction Policy |
Legal Disclaimers |
Contact Us |
My Account |
Home
|
|||||
|
|
|
|
|
||
Do you have any opinion on the differences between leading brokerages like Fidelity, Schwab and TD Ameritrade?
In truth, we don't. From our perspective, they are all top tier firms who compete very aggressively with each other to provide superior experience to their clients. We know people who work at all of them, and if they are any indication, all of these firms attract first class talent.
Do you feel your allocation models are superior to those found on other sites that are based on modern portfolio theory/mean variance optimization?
We do. As we have repeatedly noted, we do not believe that markets are generally in equilibrium and securities are priced close to their fundamental values. We believe that financial markets are filled with positive feedback loops and nonlinear effects caused by the interaction of competing strategies (for example, value, momentum, and passive approaches) and underlying decisions made by investors with imperfect information and limited cognitive capacities who are often pressed for time, affected by emotions, and subject to the influence of other people. As a result, while attracted to equilibrium, financial markets never reach it and can sometimes generate substantial over and undervaluations. In addition, complex adaptive financial markets will tend to pass through different periods during which the probability of making accurate forecasts rises and falls. We also believe that human beings have widely varying capacities for understanding the dynamics of complex adaptive systems. Finally, we are highly conscious of the limitations of all quantitative modeling approaches, and recognize that confidence in any solution increases when it can be arrived at using very different methodologies. All of these beliefs contribute to our approach to asset allocation, which is grounded in (a) the belief that asset allocation is almost always a multiperiod problem, for which single period techniques like MVO are suboptimal; (b) the use of multiple regimes characterized by very different asset class returns, risks and correlations; (c) the use of shortfall as our primary risk constraint; (d) the use of broad asset class definitions, shrinkage estimators and constraints to limit the impact of estimation errors; (e) a belief that in complex adaptive systems one must search for solutions that are robust - that have a high probability of achieving a long term portfolio return goal under a wide range of scenarios - because it is impossible to identify a single optimal solution; (f) belief that successful active management is extremely rare, but possible, and, mathematically, most valuable to a portfolio when it delivers uncorrelated alpha at a relatively low price; a (g) belief that any departure from a portfolio equally weighted across asset classes - the zero intelligence portfolio - must be justifiable both quantitative and qualitative (i.e., "plain English") terms; and (h) a belief that, because of the mathematical importance (when it comes to achieving target long term returns) of avoiding steep losses, and the inevitability of substantial overvaluations, risk management requires the constant monitoring of asset class valuations, and a willingness to occasionally go beyond rebalancing and employ more active hedging measures, like moving to cash or buying insurance (e.g., put options). Overall, we take a considerably different approach than the traditional buy and hold based on MVO (using a single set of asset class inputs derived from historical data) with regular rebalancing. While reasonable people can and do disagree about the relative merits of the two approaches (and others), we believe that research has conclusively shown that an investor or adviser will improve his or her foresight by taking all of these approaches into account when making his or her investment decisions.
Can you clarify your use of the terms "possible", "likely" and "probable" in your asset class valuation updates? Also, is your judgment about commodities valuations based on the long/short fund (LSC) or a long-only fund?
Research in the intelligence community has shown that it is far more useful to the recipient of advice to know not only an analyst's conclusion, but also his or her degree of confidence in it. In line with this research, we have take the view that, when it comes to estimating results produced by the actions of a complex adaptive system (like financial markets), quantitative confidence estimates suggest a degree of precision that simply isn't possible, and can create a false sense of security with respect to a conclusion. Hence, we have chosen to use a graduated scale of qualitative confidence indicators, with the lowest degree of confidence being "possible", followed by "likely" and then "probable". These estimates are derived from the combination of our fundamental analysis of asset class valuations, and our views about the probability of different scenarios developing in the medium term.
With respect to commodities, our valuation estimate is for a long-only position in a typical futures-based commodity index fund. However, as we have noted in the past, when it comes to implementing our model portfolios' allocation to commodities as an asset class, we concluded that the new long/short fund (of which more are in registration) seemed preferable to a long-only fund because of the structural imbalance in the market between buyers and sellers of futures contracts. In our view, this imbalance has made contangoed positions more likely, in which roll-return based profits are earned by being a futures contract seller, not buyer.