How Sentiment Can Be A Paradigm For The Markets

|
Includes: FXE, SPY, UDN, UUP
by: Philipe Saroyan
Summary

Often seen as a contrarian indicator, sentiment can also be coinciding.

A current snapshot of 5-Day Sentiment Scores for S&P 500 Futures.

Sample backtests and research support explaining trade methodology in for sentiment as an overall strategy to make money.

For sentiment to get in the ring at this stage of market activity is most definitely exciting, however it may become dangerous to view the markets this way when there is something other than a random walk taking root. With tip-of-the-iceberg type of action, peaking equity markets are the main thing. For sentiment to be a competing factor, withstanding all the real, endogenous risk factors, there has to be some liquidity that is attributed to sentiment for these indicators to work.

One sentiment indicator, which is seen as a contrarian indicator is the AAII Investor Sentiment index. This survey of professional investors gauges their positivity towards financial markets. The higher this reading the more of a contrarian indication of a price reversal is, and vice versa.

Here is a current snapshot of those sentiment readings:

In a previous study I showed how social media and options sentiment could predict daily returns of the Euro. Taking a look at the sentiment for the S&P 500 Futures from SMA data analytics - the very same data source for the research-we see the most recent sentiment scores based on their S-Factor scores. These S scores are standardized sentiment readings with mean equal to zero.

In the study, any time the 5-Day S-Factor score ticked above zero, that would indicate a buy order, and any time the 5-Day S-Factor score went below zero, that would generate a sell order. It functions as a coinciding indicator.

Taking a look at the current market snapshot, here's the S-Factor scores for the S&P 500:

With the previous couple days at negative readings of the 5-Day Average for S-Scores, there is a daily sell signal for the S&P 500 (SPX), for the following trading day. So yesterday would produce a sell signal for today, and so forth for the next trading day given a 5-Day S-Factor score of -0.682 on Monday.

Here are the results from a backtest from previous research on the Euro ETF (FXE) that included this trading strategy, labeled "5-Day Raw Score".

Notice, in the study the results for the trading strategy on the Euro had the highest overall risk-adjusted returns with a Sharpe ratio of 1.5, compared to just trading the Euro ETF FXE.

Two things should be said about this at a glance: First the Euro ETF FXE was in an upmarket for the sample backtest. Secondly, the trading strategy given 5-Day Sentiment, in this case the 5-Day Raw Score, provided by SMA produced a winning percentage greater than 50% with 257 winning trades out of a sample set of 477 paper trades giving it a so-called 'hit' rate of 53.88%, think of this as your batting average for slugging out winning trades.

With that said, we can say to sell the S&P 500 so long as the 5-Day S-Factor score remains negative.

Although the study was done on the $EURUSD and $FXE, this sentiment proxy for markets works as an add-on indicator.

Sentiment Overall

In a market where global growth comes in better-than-expected and consumer sentiment are reading better-than-expected, you can be safe conjecture that sentiment will remain a viable factor for withstanding extra profits in the markets.

Now is a perfect time to have sentiment as a paradigm for managing your portfolios. With regard to speculative price action, if the markets are at a perceived top, there should be impetus for further gains. It's still important to have a model for knowing price levels that are apparently speculative, versus true support and resistance levels.

In other words, sentiment can be of import when knowing how to cull gains and differentiating what is true price action, versus what are redundancies (e.g. media effect, path dependency, etc.)

This is especially true when there is a hyper-focus on so-called 'fake' news. Also, of intrigue is how it implies a greater potential for speculative gains, given the formula for epsilon, which is the standard error for the regression line.

The standard error formula in the Fama-French Asset Pricing Model, when factoring the Social Media Factor as a proxy for sentiment, looks like this:

Where epsilon equals the beta factor-variable "q" in this instance-multiplied by our sentiment proxy for a whole universe of stocks, in addition to profits on a population set of stocks, which theoretically should be locked in as gains to our portfolios.

Understand, this does not encompass expectations of price returns, which is one of the assumptions in modern portfolio theory (MPT). It should be pointed out, that sentiment has the pernicious quality of adjusting biases falsely when looking at sentiment as a variable in modern portfolio theory.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.