November 28th, 2011
As you know, we strongly espouse the analysis of ‘liquidity’ into one’s asset allocation modelling. This is a fairly unique approach of our software. At a recent presentation in London, Mark Carney (current governor of Bank of Canada) gave some remarks on liquidity. As a brief bio, he is a former Harvard student, former Goldman banker and currently ranked by the Financial Times as a ‘top figure in the financial world’.
His wiki entry: http://en.wikipedia.org/wiki/Mark_Carney
Global liquidity is an amorphous concept. The Usual Suspect for any event or dynamic too complicated to explain, global liquidity is the Keyser Söze of international finance. It has no agreed definition and, as a consequence, there has been no coherent policy approach to tame its more violent tendencies.
However here we have a quote which clearly emphasizes why smart money should continue to focus on liquidity’s impact on portfolio construction:
With increasing financial integration, the impact of global liquidity on domestic financial and economic conditions is growing.3 The recent Irish experience demonstrates how it can amplify the cyclical dynamics of domestic credit and asset prices.
The link to the full talk is: http://www.bankofcanada.ca/2011/11/speeches/global-liquidity/
November 20th, 2011
We are working hard on adding a ‘model validation’ section … this is where users will be able to do an array of analytics on the risk indexes they built. The major issue we may see is that if the output doesnt demonstrate some phenomenal outcome: “wow, my US Equities risk index kept me out of all drawdowns and perfectly timed the bottoms in 2003 and 2009″ then the user would interpret the outcome confusing.
There are many ways to evaluate. Lets run through some we are considering just for ‘sanity checks’.
1) risk index positive / negative: For example, taking a look at average 1-month, 3-, 6-, 12- and 24- performance of the market index, then bifurcating the returns to when the risk index was in ‘high’ risk or ‘low’ risk territory.
2) risk index at an extreme: If the risk index hit a local high (say within the past 2-3 years), what was the market index’s subsequent returns. This was the main emphasis of Covert Analytics, as a ‘risk management software’ that would help align risk index peaks / extremes with market tops and bottoms.
3) regression and correlation of risk index level with subsequent market return: A little too academic, but worth a look
4) performance of risk index as a rebalancing indicator: here we can add a binary analysis, ie if market is in high risk territory, be in cash, or if in low risk territory be in the market. The major issue here is surely the risk index will cause the portfolio to be out of the market too long, thus missing out on many many months of compounding returns.
Oh the challenges. But this will be an invigorating part of the site, that will help quantify the usefulness of our approach and software for our users.
