Research Case 1: Seasonal System for Thrift Savings Plan

Overview of System

Previous research in seasonal patterns has documented that stocks fluctuate in a fairly consistent and predictable cycle.  Over time this history does not always repeat, but it does tend to rhyme.  For instance, studies show that investing in stocks from November 1st through April 30th each year and then switching to fixed income for the other six months has produced reliable returns with reduced risk since 1950.   November, December, January, March, and April have historically been the top-performing months since 1950.  

The Research Question 

Can a system that uses seasonal trends generate greater consistent returns than buy and hold with less risk over the same time frame?  

Rules

This research case will follow a simple formula of seasonal moves that looks at a set of Thrift Savings Plan (TSP) funds that have the best historical performance for that month, and if the momentum indicators confirm an uptrend, the system places the system in that fund for the duration of the month or until the system triggers a sell signal.  If the seasonal hypothesis is valid, then the system should generate consistent reliable trading signals.  

Trackers Used

  • TSP C Fund – Will use the S&P 500 for technical analysis tracking purposes, but actual returns will be tracked from the returns of the actual TSP C Fund as posted on www.TSP.gov
  • TSP S Fund – Will use the Wilshire 4500 Completion Index for technical analysis tracking purposes, but actual returns will be tracked from the returns of the actual TSP S Fund as posted on www.TSP.gov
  • TSP I Fund – Will use the iShares MSCI EAFE ETF for technical analysis tracking purposes, but actual returns will be tracked from the returns of the actual TSP I Fund as posted on www.TSP.gov
  • TSP G Fund – Represents a move to safety (cash), so no technical analysis is required, but actual returns will be tracked from the returns of the actual TSP G Fund as posted on www.TSP.gov
  • The system will be compared against a buy-and-hold strategy of the S&P 500 Index

Variables:

  • The Halloween Indicator:  The probability of positive returns increases for stocks in Autumn, typically around Halloween.   Analysis by Bouman and Jacobsen (2002) shows that the effect has indeed occurred in 36 out of 37 countries examined, and since the 17th century (1694).  No one can explain why it works
  • Sell in May and Go Away: The “Sell in May and go away” phenomenon has persisted as a profitable market-timing strategy for stock investors.  Seasonal investing that has adhered to this rule has historically outperformed buy-and-hold by a couple of percentage points annually  
  • Momentum Indicators: In up-trending markets, momentum indicators can get a system into the market earlier and keep the system producing gains for longer.  When markets are trending down, entries are delayed until the markets turn up, and exit points come earlier to miss drops.  In past studies, when combined with another indicator, like the Moving Average Convergence Divergence (MACD), this variable has reportedly beat buy-and-hold by 5%
  • The status of Traffic Light
  • The status of 20-month simple moving average 
  • The status of 50/200-day moving averages
  • The direction simple moving averages are trending (up, down, or sideways)
    • For the test system, when the 50-day moving average is above the 200-day moving average, and the Pring KST is positive all an asset allocation change triggered into the month’s preferred fund is 100% 
    • If the 50-day moving average is trading below the 200-day moving average, and the Pring KST is positive, only make the asset allocation change 50% of the month’s preferred fund, and 50% G Fund
  • If the 50-day is below the 200-day and the Pring KST is negative for the month’s preferred fund, stay 100% G Fund for the month. Once in the investment, a sell signal would be triggered by a combination of:
    • Pring KST negative
    • Violation of trend lines and Fib levels
    • Significant candlesticks

Momentum Indicators

  • Use the Pring Know Sure Thing (Pring KST) momentum indicator to better time trades
  • Markets trend: up, down, or sideways
    • In up-trending markets, a positive KST on a seasonal day is a “buy-signal” variable
    • If the trend is down or sideways, a negative KST on a seasonal day is a “delay-signal” variable until the trend is up
  • Use Daily Charts and start reviewing 2 weeks prior to entry date.
  • If the Pring KST is positive on a seasonal day, then a “BUY” signal is generated.
  • If the Pring KST is negative on a seasonal day, then “HOLD” until the first day a “BUY” signal is generated.
  • A “BUY” day is a day to go into equities.
  • A “SELL” day is a day to go into exit equities.

Currently projected asset allocations for 2020 are provided to members in the weekly newsletter.

Recap of Seasonal Philosophy Behind Research Test Case

Simplicity: Requires only 2-8 trades a year and has produced reliable returns with reduced risk since 1950.

  • The Best Months: Historically returns are higher between November through April of each year
    • Of these six months, February is the weakest and December is the best
    • Other excellent months include November, January, April, and July
  • The Worst Months: Historically the months of May, June, and September have been bad months for equities
  • September is the absolute worse month for equities historically. In fact, if a buy-and-holder simply incorporated the method of selling equities at the end of August and re-purchasing them again in October he or she would increase their annual return by over 1%. Over time, with the power of compounding, this makes a huge difference in the size of one’s portfolio
    • The returns improve drastically when the Pring KST is applied
    • Holding time = weeks to months

About the Know Sure Thing (KST)

Developed by Martin Pring, Know Sure Thing (KST) is a momentum oscillator based on the smoothed rate-of-change for four different timeframes.  In short, KST measures price momentum for four different price cycles.  It can be used just like any momentum oscillator.  Chartists can look for divergences, overbought/oversold readings, signal line crossovers, and centerline crossovers.  Pring frequently applied trend lines to KST. Although trend line signals do not occur often, Pring notes that such breaks reinforce signal line crossovers. 

Interpretation

KST fluctuates above/below the zero-line. At its most basic, momentum favors the bulls when KST is positive and the bears when KST is negative. A positive reading means the weighted and smoothed rate-of-change values are mostly positive and prices are moving higher. A negative reading indicates that prices are moving lower. 

After basic centerline crossovers, chartists can look for signal line crossovers and gauge the general direction.  KST is generally rising when above its signal line and falling when below its signal line. A rising and negative KST line indicates that downside momentum is waning. Conversely, a falling and positive KST line indicate that upside momentum is waning. 

Even though there are many different signals possible with KST, the basic centerline and signal line crossovers are usually the most robust. Unlike RSI and the Stochastic Oscillator, KST does not have upper or lower limits. This makes it relatively ill-suited for overbought and oversold signals. 

Divergences

Bullish and bearish divergences are also possible for signals, but chartists need to be selective when using these.  Most divergences in the basic rate-of-change indicator do not result in price reversals. Similarly, divergences in MACD and RSI are also prone to failure. It is probably best to use divergences when there is a large and blatant divergence. The example below shows BroadCom (BRCM) with a large bearish divergence and a large bullish divergence. These divergences were finalized with subsequent signal line crossovers (red and green arrows). 

Strong Trends

Chartists should be careful with bearish signal line crossovers in strong uptrends and bullish signal line crossovers in strong downtrends.  KST can move into positive territory and remain in positive territory for an extended period during a strong uptrend. The indicator will reach a relatively high level and then turn down but never move into negative territory. This simply signals that upside momentum is slowing. Upside momentum is still stronger than downside momentum, but upside momentum is not as strong as in previous periods. The example below shows Sherwin Williams (SHW) with a strong uptrend from November 2011 to August 2012. Even though KST fluctuated up and down, it never broke below zero and remained in positive territory the entire time. The bearish signal line crossovers simply indicated a slowing in upside momentum, not a trend change. 

Conclusion

The KST is not a perfect indicator.  There is no such thing.  It does, however, have its uses, and can be utilized like other unbound momentum oscillators, such as MACD, the Percent Price Oscillator, and TRIX.  Because it is unbound, KST is not well suited for identifying overbought and oversold conditions.  Martin Pring, the creator, favored signal line crossovers and trend line breaks for signals. As with all indicators, KST should be used in combination with other analysis techniques and should be analyzed on all time frames (daily, weekly, monthly).

This is a private site for members only. If you are interested in observing this research in real-time, contact us by clicking here.

Disclaimer

IMPORTANT!!!!  Notice & Disclaimer about any and all content I create, send, or share with anyone, in any form regarding this research:  

This is research.  This is not investment advice.  I cannot be any clearer about this.  I am researching these systems to document their performance in real-time over a period of years for further research.  Be aware that while these systems are commonly referred to in the media, the strategies involved may not be suitable for your situation or risk tolerance.  I am not offering YOU investment advice.  I am merely conducting research on the viability of these strategies and documenting them for future research.

I will not be liable for any loss or damage whatsoever (including human or computer error, negligent or otherwise, or incidental or consequential loss or damage) arising out of or in connection with any use or reliance on the information derived from this research.  The previous statement is here not only because the lawyers insisted, but because it is true.  Consider that I have no idea what your financial goals may be, I am unfamiliar with your risk tolerances, I do not know your personal income, age, savings, financial circumstances, dependents, net worth, what tax bracket you are in.  Heck, I do not even know where you live in.  There are some people who would find these things quite relevant.

I am merely researching alternative options for viewing the markets and documenting it for further research.