Strategy Development: SPXEPS - Can EPS Predict Bull and Bear Markets?

Strategy Development: SPXEPS - Can EPS Predict Bull and Bear Markets? Header Image

Some portions this post contains premium content, which are marked, and can only be viewed by premium subscribers. To view the whole post sign up or review pricing.

This post I continue, that we started earlier from this post, to refine the long-term trend trading system that uses the EPS of the $SPX to define buy an exit points from 1998 to today. I also look for some additional correlation across other asset classes such as $GOLD, $30Y bonds, and real and nominal GDP. The main purpose of this strategy is to help determine broad market trends and trend changes, but also show how you can also refine your own trading strategies.

In my previous blog post when sentiment in early 2019 was bearish on a -20% decline in the market I suggested that market sentiment could be getting ahead of itself based on the fact that we hit a new all-time high in !GAAPSPX earnings. I painted in broad strokes of potential strategy that gives us buy and sell indicators based off changes in EPS increases or decreases.

Let’s establish a name for the strategy:

SPXEPS

This post helps to mold the hypothesis into something more concrete and help refine SPXEPS as well as test it against some other metrics when I do get a sell signal. Also in a future blog post I will evaluate a money management system.

Strategy Summary:

Difficultly of strategy:

Beginner

Symbols:

  • $SPX (S&P 500),
  • /GC or $GOLD (CME Gold Spot USD)
  • /ZB (CME 30-Year Bond Futures)

Indicators:

  • $SPX !GAAPEPS

Buy Signal (Cover if short):

  • Default position is to buy or be long $SPX
  • Coming from any period where we are in a sell zone, when we turn from a negative to positive EPS (any positive EPS amount works)
    • Example we go from -1.00 EPS to 0.01 EPS this constitutes a BUY signal

Sell Signal (Short to open):

  • A <= -3% EPS change from QoQ or data point to data point

Stop %:

  • -5% for long $SPX positions
  • -7% for short $SPX positions
  • -3% for $GOLD positions

Target %:

Not defined for any system.

I like this strategy because we are looking at a number related to the SPX but not directly TIED to the SPX. For instance many moving averages could be used to account for trend following strategies of the SPX that can give us both buy and sell signals, but $SPX !GAAPEPS is not directly tied to the price chart. It is a fundamental number, not a technical number. Later we will also look at $SPX Revenue per share and I will give some thoughts on that.

Results:

Link to full sheet is Premium Content

Inferences:

  • Long $SPX as a default position is a decent idea. The default position should always be to buy or looking to buy. Finding reasons not to buy seems to be prudent.
  • Sell signals are triggered when we have an EPS contraction of -3% or greater from data point to data point.
    • In the early 2000s the charts were less refined. I imagine what was happening is as companies were reporting earnings adjustments were made. Overall and throughout time the data was smoothed and finalized to take these adjustments into account. Reporting was also streamlined as indexes were refined themselves, quantified, and decimalized.
    • I also removed all data points that were less than +/- 0.5% change, as I considered them “noise” and immaterial to the strategy’s trade signals and rolled the percent change into the next data series. This did NOT have any material impact on the data.
  • Long gold only since 1998 was a better performance, 347%, from the spot $SPX and $SPY (dividend’s reinvested). $SPY long, which includes reinvesting dividends was at 269%, while $SPX spot was at 159% increase. Meaning a large amount of gains of the market are because of dividends, which come from earnings (and are strongly influenced by interest rates).
  • Long gold during times of sell signal instead of shorting $SPX yielded a higher success rate (100% vs. 50%). However the caveat is in the details when gold has had a lower R value of 0.67 & 0.96 in the last two SPXEPS sell signals. Unrealized max drawdowns were twice as high as short $SPX trades on average. Gold was therefore more profitable to go long during these periods, but also carried more risk.

I go in detail in the video below, walking through almost every aspect of the data.

The first video is for all public viewers, the next videos are for premium subscribers.

Part 1 of 2

In this video I review the EPS and Price data in helping prove or disprove the hypothesis of EPS can be used to predict bull and bear market moves of the broad market or the S&P 500.

Part 2 of 2 (1.5 hrs)

After looking at $SPX EPS, Real and Nominal GDP, Gold, and Bonds, I determine the trade tables and give an approach on how we can continue to the next part of the strategy development which is money management.

Premium Content

Now I have identified some of the fundamentals of the SPXEPS, I can now begin to overlay a money management system. I can a money management system to the trend following trade strategy by using stock ETFs, options, and futures.

Final Thoughts:

SPXEPS appears to be a decent trend following strategy; further research can determine if it meets my personal criteria on profitability, and can it fit into a money management system so it matches my trading personality.


Recent Posts


Sign up for the Newsletter

Get email alerts when a new blog is posted, or a public strategy trade is made

Comments