Leif Peterson's  Instablog

Introduction

The wealth of a nation strongly depends on its economic conditions.  At present, most governments (households) don't save money, so in order to grow, additional debt must be taken on in the form of overleveraging. It can be argued that the US economy and the dollar are headed for failure as a result of an unbalanced budget, excessive debt in the form of US Treasuries, growing taxation, a growing trade deficit, and decreasing GDP. The formal definition of leverage is the ratio of total assets to assets minus debt.  As an economy improves, asset prices improve, leverage falls, corporate balance sheets increase in order to use excess capacity, and new liability is taken on in the form of short-term debt.  A circumlocution begins in which the "dog chases its tail," giving rise to a cyclical bull and bear market fluctuations of equity price returns.

There is tremendous uncertainty in the cause of bull market bubble-bursts, ranging from devaluation of fiat currencies, within-sector corrections that affect broad markets, to decoupling of valuation. The additional volatility that ensues increases with the risk of each investment, requiring more time to review performance and identify successful growth strategies.  During the search for factors predictive of growth, the priority becomes drilling down into the deeper levels of information and minimizing strategies for identifying global patterns predictive of long-term bull markets.  Although prediction of future price returns has proven difficult, there is merit in the use of moving averages and momentum indicators for prediction of long-term market patterns.

Moving averages of daily price returns can be combined to compare current investor expectation with average expectation over a longer period.  A bullish behavior of buying equities is evident when a shorter moving average exceeds a longer-term moving average.  Analogously, a bearish selling behavior is indicated when the difference between the short- and long-term moving averages is negative.  The majority of momentum indicators in use today work best when applied to within-stock time series for a specific stock.  Correlations of lagged returns with other market indices have proven to be less predictive of future returns when compared to within-stock features and optimal signals.  With the recent increased volatility witnessed throughout the subprime mortgage crisis, and sideways trending markets expected over the next several years, there is growing interest in successful prediction models that adjust to the nuances of a particular stock.  For this reason, the introduction of global pattern indicators for judging when to enter or exit the global markets occurs less frequently. The intent of this article is to define a global market indicator called "Get Into The Market Oscillator [GITMO]", to be used as an entry/exit signal for prioritizing investment in equities.  Awareness and use of the GITMO oscillators could possibly help investors know when to get into the market and invest in equities more heavily and transition out of asset preservation during long periods of economic downturn.  On the contrary, the GITMO oscillators can signal when to leave the equity markets and prioritize asset preservation.  The next section describes the methods used for calculating the GITMO oscillator, which is followed by conclusions. 

Methods    

Five GITMO oscillators are introduced, ranging from short-term to long-term (i.e., 1, 2,...,5).  The generic fundamental equation for GITMO is

GITMO = min{constant, exp[zero-lag EMA(t1) - zero-lag EMA(t2)]},     t1 < t2

where the constant is a fixed value from the set {60, 70, 80, 80, 100}, EMA is the exponential moving average of the S&P 500 price return index for the historical time periods t1 and t2. The constraint that t1 < t2 forces t1 to be a shorter time period than t2.  (At Yahoo Finance, the price return index of the S&P 500 is ticker ^GSPC, while at Google Finance the ticker is .INX). The use of zero-lag EMA helps minimize the chance for missed opportunities after rapid short-term growth spikes.  The five GITMOs are defined as follows:

GITMO-5 = min{100, exp[zero-lag EMA(320)-zero-lag EMA(500)]}

GITMO-4 = min{90, exp[zero-lag EMA(250)-zero-lag EMA(500)]}

GITMO-3 = min{80, exp[zero-lag EMA(120)-zero-lag EMA(420)]}

GITMO-2 = min{70, exp[zero-lag EMA(120)-zero-lag EMA(250)]}

GITMO-1 = min{60, exp[zero-lag EMA(25)-zero-lag EMA(120)]}

GITMOs were intended to be discrete rather than quantitative. By design, a unique property of the GITMOs is that when the difference of the shorter and longer moving average is negative, the exp[] term falls in the range [0,1], otherwise, positive differences cause exp[] to range from 1 to positive infinity.  Large values of exp[] are dampened by the constant term, so that GITMO purposely reflects its fixed discretized values of 60, 70, 80, 90, or 100 depending on which GITMO is considered.  GITMOs were developed for the daily price returns of the S&P 500 index, but can be applied to any stock.  The rationale for using the S&P 500 price return index as a basis for GITMOs is that the S&P 500 is the weighted price return of 500 large cap common stocks of US industries traded in the US.  That other indices of the US markets correlate with the S&P 500, GITMOs are useful for determining long-term periods of overall equity market growth and contraction, ultimately portraying current economic conditions. 

Several properties of GITMOs have been observed.  For a long-term projection on when to prioritize investment in equities, the GITMO-5 is recommended, since it incorporates a 500-day moving average.  The figure below illustrates the GITMO-5 of the S&P 500 over the last 15 years.  

As one can note, when the maximum constant value of GITMO-5=100 was reached, the S&P 500 incurred strong periods of growth during 1995-2000 and 2003-2007.  In January 2008, GITMO-5 approached its near-zero values.  At this point, when GITMO-5 drops to its low near-zero value, investors might transition to inverse ETFs for the S&P 500, sharpen their focus to identify individual equities that are nevertheless rising in price during the downturn, and go bargain hunting.  Moreover, people who don't follow their mutual funds or employees whose retirement accounts are limited to mutual funds only might prioritize transition to asset preservation mutual funds which invest in short-term Treasuries, cash reserves, money market CDs, or gold and precious metals.  One thing is certain about the GITMO-5: when it drops from it maximum to near-zero values, it will likely take 2-3 years before long-term growth of the S&P 500 is observed, assuming the past is predictive of the future.  If GITMO-5 intermittently spikes and hits 100 for only short periods of time, then volatility will be the current landscape, and investors may want to seek conservative low-risk or less volatile investments in order to avoid whipsaws.  The choice of making a move to safer investment vehicles depends on the goals of the investor and the portfolio being considered.  For anyone nearing retirement during a period when GITMO-5 is not at its maximum value of 100, it would be advisable to preserve assets and possibly harvest dividends from strong companies whose earnings and dividend payouts are increasing.   

The GITMO-3 and GITMO-4 oscillators will commonly rise several months prior to GITMO-5 as observed in the next two figures for 1995 and 2003.  In 1995 and 2003, GITMO-4 reach its maximum of 90 about 2 months after GITMO-3 reached its maximum of 80, which was followed by GITMO-5 reaching its maximum one month later.  Historically, this characteristic sequential pattern among GITMOs 3,4, and 5 in 1995 and 2003 occurred at the beginning of long-term bull markets for which the average growth in equity price returns was the greatest.   

This year, in 2009 (figure below), GITMO-1 reached its maximum on March 19, followed by a maximum of GITMO-2 on May 28, and then maximum of GITMO-3 on July 30.  Again, this pattern of maximum GITMO values may be suggestive of the beginning of another long-term bull market among the S&P 500 similar to 1995 and 2003.  Assuming the pattern of consecutive maxima being reached for GITMO-3, GITMO-4, and GITMO-5 in 1995 and 2003, there is probably a good chance that the maximum GITMO-4 value will be reached at the end of September or early October, followed by GITMO-5 reaching its maximum near the end of October or early November.      

Conclusions

The GITMO oscillators presented in this article were developed to be discrete indicators of historical long-term growth in the price return index of the S&P 500, and to indirectly be used for entry/exit signals for prioritizing investment in equities.  Long-term increases and decreases of the S&P 500 price index have often revealed the direction in which the US economy was headed.  When the long-term GITMO-5 is at its maximum of 100, there may be added confidence that a long-term bull market is occurring.  Past behavior of GITMO over the last 15 years leads to following conclusions:

1.  Whenever GITMO-5 dropped from its constant value of 100, there was a significant 2-3 year downward momentum in the price index of S&P 500.  Hence, a drop of GITMO-5 from its constant value of 100 may be viewed as an exit signal to transition to inverse ETFs, safer investments, and asset preservation via short-term Treasuries, dividend harvest, other currencies, gold and precious metals.  This is especially important for people nearing or in retirement, people who don't track their investment performance, or employees whose retirement accounts are limited to investing only in mutual funds.      

2.  When a drop in GITMO-5 was followed by all the GITMO oscillators reaching their maxima sequentially in the order 1,2,3,4, and 5, it signaled the beginning of a long-term growth period. This pattern could be considered as an entry signal for prioritizing equity investment and transitioning out of asset preservation.  Investors with access to only mutual funds may wish to begin long-term investing in mutual funds that invest in growth stocks.  Whatever is done, there may be a lower priority for asset preservation during this period.    

3.  GITMO-1 was too jumpy and was not nearly as informative for identifying long term entry and exit signals when compared with GITMOs 2-5.  

4.  GITMOs for the S&P 500 were not intended to reveal price return characteristics of specific sectors, commodities, or currency markets.  The future performance behavior of GITMOs for the S&P 500 will be affected by interest rates, inflation, leverage, liquidity, taxation, unemployment, and political events which have a major influence on the US economy and equity markets. 

The huge losses suffered by investors who didn’t know when to get out the equity markets during 2007-2008 could possibly be minimized by use of GITMO if similar patterns in the S&P 500 occur again in the future.  For employees with restricted access to only mutual funds in their employer retirement accounts, there is no reason why they had to continue to invest in the same equity-based funds -- especially when an asset preservation strategy could have minimized their losses.  GITMO was introduced for judging when to enter the equity markets and leave an asset preservation strategy, or exit the equity markets and initiate asset preservation methods.        

Disclosures: this article is not intended to serve as investment advice, but rather to introduce the GITMO oscillators for judging when to get into the equity markets after sitting on the sidelines or prioritizing asset preservation during market downturns.