We go against the trend in this article and show a bullish or at least not bearish chart.
After a long and persisting uptrend, technical analysts have run out of ammunition as most indicators that traditionally signaled major tops or bear markets have failed consistently. Lately there is a shift to debt charts maybe out of desperation. Debt dynamics are slow and charts based on indicators derived from debt are not effective in establishing major turning points. In other words:
Debt levels can increase for longer than shorts can remain solvent.
We often go against the herd and look for information that falsifies or at least counters their thesis. The information must be provided by a relatively simple indicator otherwise one risks becoming the victim of complexity and over-fitting.
Below is a “simple” indicator that is based on the mean and variance of log daily returns in a lookback period N. This is also known as the Kelly Optimal Leverage, originally developed by legendary quant Edward O. Thorp for optimal continuous rebalancing.
The indicator is equal to the expected excess return in a time period divided by the expected variance of returns in the same period.
f = μ/σ2
where f is the optimal leverage for continuous rebalancing, μ is the excess return and σ is the standard deviation. (This is not to be confused with the Kelly ratio used for optimal position sizing.)
Below is the indicator applied to S&P 500 index for 500-days lookback period.
The current value of the indicator is a little below the longer-term average since 1960 and much below values reached before the market tops in 1960s, 1970s, 2000s, the 2016-2016 corrections and even the more recent 2018 correction.
It is also important to notice that high values of the indicatorhave occurred near the start of long uptrends, for example in 1996 and 2004.
If we increase the lookback period to 1000 days, or approximately four years, this is what we get.
The current value of the optimal leverage is a little above the longer-term average but nowhere close to values realized before the 2000 and 2007 tops and even before the 1987 crash.
The conclusion is that this quantitative indicator does not forecast a top at this point and in fact it hints at continuation of the uptrend.
However, indicators always lag price since they are derivatives and more importantly do not reflect any form of causality. Therefore, they must be used with caution.
Our objective here was to present an indicator that offers a different picture than the numerous indicators presented in financial blogosphere lately that presumably forecast an end to the stock market uptrend.
If you found this article interesting, I invite you follow this blog via any of the methods below.
If you have any questions or comments, happy to connect on Twitter: @mikeharrisNY
Charting and backtesting program: Amibroker