The Conceptual Link Behind the “LTR” Ranking

“Does positive feedback trading, indicated by an adjusted measure of return autocorrelation, enhance momentum profitability? In the February 2010 version of their paper entitled “Positive Feedback Trading Activities and Momentum Profits”, Thomas Chiang, Xiaoli Liang and Jian Shi examine the relationship between positive feedback trading and profitability of momentum strategies.”

Readers are encouraged to look at  the following link on CXO Advisory as important background reading: http://www.cxoadvisory.com/volatility-effects/amplifying-momentum-returns-with-idiosyncratic-volatility/

I have often been asked whether the Livermore Index is not simply an index of “high-beta” stocks that do well in up markets and poorly in down markets. The  hedged backtests show a sharpe ratio of 2 across bull and bear markets, so therefore it would be mathematically impossible for this to be the case. So what is driving the alpha above and beyond standard relative strength excess returns? The real reason the Livermore Index has excess alpha is due to is “LTR” ranking, which is a proprietary measurement of autocorrelation—and not the standard variety. https://cssanalytics.wordpress.com/2010/04/06/performance-of-dv2-on-the-bottom-50-ltr-dv-index-versus-the-top-50-ltr-trend-index/ The key is that stocks with high LTR rankings are much more likely to trend smoothly and have performance that is less connected to the index. In fact, such stocks exhibit high idiosyncratic volatility versus beta volatility–which is a fancy way of saying that their volatility is due to individual business factors versus index movements. Idiosyncratic volatility provides two distinct benefits 1) it improves the diversification benefit dramatically since correlations are lower between stocks with high idiosyncratic risk versus low idiosyncratic risk 2) it permits a stock to detach from the index.

This means that stocks that have powerful fundamentals and a strong business model like an Apple Computer or a Baidu.com will be able to march to the beat of their own drummer at times, rising even when the market may be falling. This detached behavior is especially evident on the shorter time frames. Since high LTR stocks must exhibit a smooth price history, the detachment often manifests itself in a persistent/consistent manner. This behavior helps to improve the risk-adjusted returns of holding such stocks since they tend to have strong momentum versus volatility.

But what about the reverse? I introduced the low-LTR index many posts back as a “mean-reversion” index. The differences in the performance between the high and low LTR indices was remarkable– high LTR stocks often lost money trading using mean-reversion strategies while low LTR stocks made a significant profit. So what is driving this relationship? For one, low LTR stocks are very choppy and tend to have less momentum. This means that they often have more support and resistance barriers, and thus tend to stay range-bound. Low LTR stocks are also more volatile proportionately, and as we know volatility tends to be cyclic.  But finally, and perhaps most importantly, low LTR stocks are more likely to exhibit low idiosyncratic volatility and therefore by extension tend to have a higher beta to the index than high LTR stocks. A higher beta implies that low LTR stocks will be more in sync with the index. Thus if they become over or under-valued versus the index by way of being overbought or oversold, they will “snap back” due to their higher cointegration relationship.

The bottom line is that despite many of the catchy monikers or acronyms, there is almost always a logical reason behind many of the ideas put forth on this blog. Many have their roots in academic finance theory. Initially it was my intention to spare readers from much of the logic to avoid making things sound overly technical or beyond reach. However, I think it is worthwhile to consider such things so that we all do not get trapped in the semantics and the fluff. The reality is that indicator and strategy names can almost make it seem as if there are more different/distinct ideas than actually exist. The truth is oscillators, trend indicators, relative strength, breadth all share a very common lineage—it is up to the smart researcher to think in terms of what they are trying to find out—or what they are trying to take advantage of versus attempting mass optimization and data mining. Often it is just easier and more tangible to proceed in such a manner–but I implore you to try to think and understand the market because that is where you will find true enduring anomalies versus spurious or unstable results.