Model With Sell In May Variable

As the upcoming actionable signaling within our quantitative tactical strategy involves the “Buy in November, Sell in May” premise ( buy equity class on November 1, sell on May 1 * ), further investigation into “Buy in November, Sell in May” combined with Market Map model variables, has uncovered some benefits that can be useful in the investment of diversified 4 stock universe portfolio.

Our research suggests the use of an equal weighted “blend” of the small cap value, emerging small cap, mid cap growth, and large cap value universes during the November 1 through April 30th investment periods ( or longer dependent on risk profile instruction ), and the utilities sector ( or bonds, cash / money market equivalent ) during the May 1 – Oct 31 investment period. Over the historical sample period from 11/01/1953 to present, risk adjusted excess returns ( alpha ) were produced over the buy & hold of the “blend”.

The allocation instruction of the strategy entails:

1) the use of a variable in determining “high” risk in forward year equity market returns ( a calculation of percent return on the S&P 500 index over a fixed date time series from the mid November to mid January ( two month ) time period with negative readings correlating to negative forward year stock market growth )

2) the use of a long length moving average cross rule calculated on the relationship between the price of the S&P 500 and it’s long length moving average

3) bond allocation from May 1 through Oct 31 using a moving average cross rule applied to bond prices * ( if items #1 and # 2 are satisfied ).

4) the use of a variable determining shorter term equity market risk from the May 1 – Oct 31 period through the calculation of 2 consecutive years of the S&P 500 > X %. On occurrence, allocation towards utilities sector from May 1 –  Oct 31 has been instructed.

These instructions lock the transaction dates to just May 1 / 1st trading day in May and November 1 / 1st trading day in Nov. The use of 4 stock universe blend helps diversify across a broader investment spectrum. Additionally and historically, there have been periods when emerging equity markets have become “undervalued” and domestic equity markets were fair to “overvalued” ( as is presently the case ). Exploiting this valuation discrepancy can provide possible dampening of the returns volatility accompanied with the “mean reverting” of valuations while simultaneously providing diversification over many markets and companies.

CAPE valuation measures, Emerging vs. U.S.


Allocation Heuristic =

1) buy any of / equal weighted blend of Small Cap Value, Emerging Small Cap, Mid Cap growth, Large Cap Value on November 1st of each year.

2) On February 1, calculate return on S&P500 from mid November to mid January ( risk profile ). If risk profile return calculation is > 0, then hold equity asset blend. If risk profile returns calculation is less than < 0 ( forward high risk ) then #3.

3) On May 1, measure price of S&P 500 index in relation to it’s long length moving average. If price is above > MA, then hold equity blend.  If price is below < MA then # 4.

4) Sell the 4 blend equity based assets and calculate the 10 period moving average, monthly basis on the Vanguard Long term Treasury fund ** (MUTF:VUSTX) . If price of VUSTX > 10 SMA, then allocate to intermediate/long term bonds / duration assets from May 1 to November 1. Otherwise, ( if price of VUSTX < 10 SMA ) allocate to cash.

5) On May 1, if  the 2 most recent consecutive annual, year end S&P 500 returns are greater than > X % ( total return ) [ proprietary] has occurred, then allocate to utilities sector from to May 1 – November 1 ( this is opposite of step # 2 & 3 as the S&P 500 would be > than it’s moving average ( the equity market is extended and vulnerable to price decline, statistically )

6) Allocate back to 4 equity asset blend on Nov 1 if # 4 or # 5 have been in force.



Concluding Points …

In order to produce optimal risk managed asset accumulation ( alpha ) in an empirical fashion, an investor can construct a systematically based investment policy using:

– a diversification of stock universes that have produced highest excess returns historically over a long sample**

– risk mitigation variables that have shown positive outcomes, statistically

– asset classes that have negatively correlated price movement or “safe” assets during periods of high risk ( long dated government bonds )




Supplemental article data here:

Some of the funds representative of the stock universes and asset classes for use in the strategy: VBR, VIOV, DFSVX, DGS, DEMSX, MDY, QQQ, PRF, FSUTX, FUTY, TLT, VUSTX

High risk + Moving average validation years = 1962, 1970, 1974, 1982, 1990, 2001, 2002, 2008

Consecutive years S&P 500 year end returns = 1956, 1965, 1977, 1981, 1984, 1987, 1997, 1998, 1999, 2000, 2011, 2014

* research has explained “seasonal” calendar year periods that have shown repeated statistical tendency towards  producing the weakest  stock market returns ( May 1 – October 31 ) and strongest returns ( November 1 – April 30 )


** data for moving average calculation – 10 year bond yield data 1954 – 1983 and VUSTX data 1983 – 2016

** Fama and French

Data sources and calculations = IFA Index Calculator, DFA funds, VUSTX, FSUTX, longrundata,

As diversification of stock universes is important, diversification of strategies is additionally important. As the “Sell in May” with Map model strategy has shown excellent results, it may not be prudent to invest all of one’s total capital towards it.

Signaling History:

Compound annual growth 1954 – 2016











Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s