1/ Want Smart Beta? Follow the Smart Money: Market and Factor Timing Using Relative Sentiment (Micaletti)
"We present a real-time, cross-asset, positions-based relative sentiment indicator derived from the COT report to predict the U.S. equity market."
papers.ssrn.com/sol3/papers.cf…
"We present a real-time, cross-asset, positions-based relative sentiment indicator derived from the COT report to predict the U.S. equity market."
papers.ssrn.com/sol3/papers.cf…
2/ "We propose a cross-asset, positions-based relative sentiment indicator derived from the COT report. It measures the aggregate degree to which institutional investors are long/short equities (+positions in closely related non-equity assets) relative to individual investors."
3/ "To get a sense of SMI's ability to time the market factor, we first plot the cumulative returns when prior-week SMI is positive/flat otherwise vs. when prior-week SMI is negative/flat otherwise.
"To demonstrate SMI's robustness, we consider 16 computational variants."
"To demonstrate SMI's robustness, we consider 16 computational variants."
4/ "Each SMI case that analyzes the full data set (Cases 1-7) has a solid, often very large, contingent of parameter combinations exhibiting high stat. & economic significance.
"Higher returns when SMI is positive come with lower volatility, contrary to what MPT would predict."
"Higher returns when SMI is positive come with lower volatility, contrary to what MPT would predict."
5/ "SMI is able to identify the times when it is advantageous to be long the market during states of negative momentum. This helps address one of the weaknesses of time-series momentum: its tendency to maintain reduced equity allocations too long after the market has bottomed."
6/ "We ran regressions of weekly excess market returns against a combination of SMI and the two momentum indicators used above.
"In the presence of SMI, time-series momentum appears to add virtually no incremental information."
"In the presence of SMI, time-series momentum appears to add virtually no incremental information."
7/ SMB and ST reversal do better when SMI is positive.
HML, RMW, CMA, LT reversal, BAB, and QMJ do better when SMI is negative.
SMI timing works better for Group 2, presumably because investors are more speculative (buying low-quality glamour stocks) when SMI is positive.
HML, RMW, CMA, LT reversal, BAB, and QMJ do better when SMI is negative.
SMI timing works better for Group 2, presumably because investors are more speculative (buying low-quality glamour stocks) when SMI is positive.
8/ The authors test a walk-forward tactical asset allocation strategy based on SMI, their Smart Money sentiment Indicator.
It has a higher Sharpe ratio along with a lower average equity allocation than the other strategies tested, including CAPE+TREND, but with higher turnover.
It has a higher Sharpe ratio along with a lower average equity allocation than the other strategies tested, including CAPE+TREND, but with higher turnover.
9/ "SMI may be a both more powerful and more efficient as a TAA tool than either value or momentum."
"SMI seems to be capturing the deep and persistent forecasting advantages that institutions have over individuals (and the time horizons over which those advantages play out)."
"SMI seems to be capturing the deep and persistent forecasting advantages that institutions have over individuals (and the time horizons over which those advantages play out)."
