Proprietary Models

We designed our public dual momentum model, GEM, to be simple and easy to use by do-it-yourself investors. GEM exists to help protect smaller investors from horrendous drawdowns while allowing them to earn better than market returns over the long run.

But dual momentum works best when it incorporates multiple trend determinants. There is a synergistic effect from doing so. Most investors do not give enough importance to price trend. But Greyserman & Kaminsky show that simple trends have outperformed buy-and-hold back to the beginnings of every market. No other investment factor can say that. 

We created our proprietary models to achieve high and consistent returns using all available investment tools. Our models are the culmination of a lifetime of investment research and experience. 

They are highly adaptive to market conditions and often serve as core portfolios. They are based on academic research and supported by out-of-sample and real-time validation.

Most investment approaches do not spend enough time on portfolio construction. Thoughtful portfolio structuring is an important part of optimal investing. We spend considerable time on due diligence to find the best ETFs for our models.

We use multiple models that complement each other to create optimal portfolios. Because of their modest correlations, our models work best when used together. Using multiple models also reduces model estimation errors and uncertainty.

We license our proprietary model signals to substantial private investors, institutional investors, and investment advisors. 

Broad-Based Proprietary Models

These use multiple dual momentum modules to hold various U.S. and non-U.S. stocks, bonds, commodities, managed futures, or gold. 

Enhanced Global Equities Momentum (E-GBM)

E-GBM is our most diversified model. It selects four ETFs at a time from a broad portfolio of potential assets that respond favorably to dual momentum. 

Dual Momentum Fixed Income (DMFI)
Dual momentum works as well with fixed income as it does with equities. DMFI is our simplest proprietary model. It applies dual momentum to the short and intermediate segments of the bond market. DMFI usually holds two ETFs. Since 1970, DMFI has had equity-like returns with the volatility of shorter-term bonds. 
Advanced Global Equities Momentum (A-GEM)

A-GEM focuses on various stock market areas but includes bonds and other assets when market conditions warrant. It adds market structure to dual momentum for determining trends. A-GEM uses daily and weekly data instead of only monthly returns and usually holds four ETFs. 

Focused Proprietary Models

These models generally use an exceptional channel breakout-type trend system applied to daily data. Our latest research shows the efficacy of this approach on 100 years of industry data, giving us added confidence in our models. Richard Dennis taught this approach to his “turtle traders.” Jack Dreyfus became a billionaire using channel breakouts of stocks making new highs. 

Most other models exit to a safe harbor asset when not in their risk-on positions. Our focused models are unique in that they switch to alternative assets and other models with positive trends before seeking a safe harbor. This layering approach captures additional profit and reduces whipsaw losses.

Our models work well in many markets. We chose the assets they use so that the combinations will create balanced portfolios adaptive to all market conditions. By doing this, they also have the highest reward-to-risk ratios.

High Yield and Municipal Model (HYMM)

HYMM applies trend filtering to the SPDR Nuveen Bloomberg High-Yield Bond ETF (HYMB). When the trend of HYMB is not positive, HYMM holds the SPDR Nuveen Bloomberg Short Term Municipal Bond ETF (SHM) if its trend is positive, or the iShares 0-3 Month Treasury Bond ETF (SGOV). Bond interest from HYMB and SGOV is exempt from U.S. income taxation.

NASDAQ Broad Advanced Trend (QBAT)

QBAT applies trend and mean reversion to the Invesco QQQ Trust (QQQ) when its trend is positive. QBAT holds the same positions as our A-GEM model when not in QQQ.

Gold Long Trend (GLTR)

GLTR applies trend following to the SPDR Gold Shares ETF (GLD). Gold is often mean-reverting and difficult to trade, but our trend strategy handles it well. GLTR holds the A-GEM positions when it is not in gold.

Bitcoin and Digital Asset Synergistic System (BADASS)

BADASS applies our trend model to the BLOK ETF, representing blockchain technology, and to a spot Bitcoin ETF. This combination gives better results than either by itself. When not in these positions, BADASS holds the GLTR positions.

Stock Market Upside Reversal Factor (SMURF)

SMURF is a stock and bond model that combines mean reversion with trend. It buys a global technology ETF (IXN) on weakness when its long-term trend is positive. SMURF exits when the trend changes direction or when there is abnormal strength. When not in IXN, SMURF  holds taxable high-yield bonds or the HYMM position, depending on their trends. 

Performance

Here are the results of the current version of our A-GEM model compared to the S&P 500, QQQ, and a 70%/30% balanced stock/bond portfolio. We also show 50/50% and 70/30% allocations to A-GEM and DMFI. Results began in January 1999 due to limited daily data before then. We have extended ETF data with associated index data when necessary.

A-GEM and DMFI Model Performances  –  September 1999 through September 2024

      S&P500   BALANCED

 QQQ

 DMFI  AGEM  50/50  70/30
CAGR        8.0             7.2    10.2     8.4     18.5    13.4    15.5
STD DEV      16.6           11.8    25.4     5.1
     11.7      7.5      9.1
SHARPE     0.55           0.64    0.51   1.56     1.52    1.71    1.63
UPI

     0.59

           0.84    0.32   5.48     7.10    8.65    8.00
MAX DD   -52.9         -38.1   -79.6    -5.0    -10.6     -6.1

    -7.9

AVG DD   -10.3           -5.4   -31.6    -0.9      -1.5    -0.8     -1.0
% UP MOS        62             64       60       74        67        73        70

 

Results do not guarantee future success nor represent returns that any investor attained. Performance includes reinvestment of interest and dividends. CAGR is the compound annual growth rate. We have estimated transaction costs to be 0.1% per trade. Balanced is 60% S&P 500 Index and 40% ICE U.S. Treasury 7-10-Year Bond Index. Maximum drawdown is on a cumulative month-end basis. UPI is the Ulcer Performance Index, which divides return by the Ulcer Index. The Ulcer Index measures the depth and duration of drawdowns from earlier highs. Please see our Disclaimer page for more information. 

U.S.-based taxable accounts may prefer to allocate their fixed-income funds to high-yield municipal bonds (HYMB). HYMM applies a trend overlay to that market when its trend is positive. Here are the results of HYMM, DMFI, HYMB, and a 50/50% allocation to DMFI and HYMM.

HYMM Model Performances – May 2011 through September 2024

 

       HYMM

        DMFI

       HYMB

      50/50     

CAGR

          7.5

          6.5

          4.7

           7.1

STD DEV

          5.0

          4.4

          7.6

           3.8

SHARPE

       1.48

        1.45

        0.63

         1.82

UPI

       6.45

        5.35

        0.82

         8.06

MAX DD

       -6.3

        -3.8

       -19.0

          -4.7

AVG DD

       -0.6

        -0.8

         -3.5

          -0.5

W% MOS

          70

          73

            69

             73

Results do not guarantee future success nor represent returns that any investor attained. Performance includes reinvestment of interest. Positions are rebalanced monthly. CAGR is the compound annual growth rate. Maximum drawdown is on a cumulative month-end basis. UPI is the Ulcer Performance Index. This divides the average return by the Ulcer Index, which measures the depth and duration of drawdowns from earlier highs. See the Disclaimer at the end of this document for more information.

 We can combine DMFI or HYMM with QBAT and GLTR to create a “barbell approach” of non-correlated low- and high-volatility strategies. Here are the back-tested results from January 1999 with different allocations to DMFI, QBAT, and GLTR in that order. Since QBAT and GLTR hold the same ETFs as A-GEM when not in QLD and GLD, these allocations represent four different models.  

DMFI, QBAT, AND GLTR Model Performances  –   Jan 1999 through September 2024

    DMFI  QBAT  GLTR  50/25/25  50/30/20   40/30/30 34/33/33
CAGR      8.4    19.2   17.9     13.6      13.7      15.4      15.3
STD DEV      5.1    14.5   14.8       7.6       7.6        8.4        8.9
SHARPE    1.56    1.27   1.19     1.72     1.72       1.67      1.64
UPI    5.48    7.56   5.09     9.67     9.47       9.38

      9.17

MAX DD    -5.0  -12.6 -13.2     -5.7      -5.9       -6.5       -7.0
AVG DD    -0.9    -1.5   -2.3     -0.7      -0.8       -0.8       -0.9
W%MOS       74      65      62        71         72          69          69

Results do not guarantee future success nor represent returns that any investor attained. You cannot invest directly in our models. CAGR is the compound annual growth rate. Drawdowns are on a month-end basis. UPI is the Ulcer Performance Index, which divides return by the Ulcer Index. The Ulcer Index measures the depth and duration of drawdowns from earlier highs. Please see the Disclaimer page for additional information.

 SMURF Performance  – May 2011 through September 2024

Instead of being broadly in stocks and bonds at the same time, SMURF is in technology stocks (INX) stocks when they have a high probability of success. SMURF spends 63% of its time invested in bonds. It combines mean reversion with trend and has the highest reward-to-risk ratio of all our models. Conservative investors may want to consider something like a 50/50% in SMURF and HYMM to achieve stock market-like returns with substantially less expected volatility.

 

       HYMM

      SPY

      INX

      SMURF

     50/50

CAGR

          7.5

      13.3

       17.5

         19.2

      13.4

STD DEV

          5.0

      15.6

      18.6

         11.1

        6.7

SHARPE

        1.47

      0.88

       0.96

         1.64

      1.90

UPI

        6.36

      2.40

       2.36

         5.64

      8.09

MAX DD

         -6.3

    -23.8

     -34.0

       -12.2

       -7.2

AVG DD

         -0.6

      -3.3

       -4.3

         -1.7

       -0.8

W% MOS

           68

         68

          66

            75

         76

Results do not guarantee future success nor represent returns that any investor attained. You cannot invest directly in our models. CAGR is the compound annual growth rate. Drawdowns are on a month-end basis. UPI is the Ulcer Performance Index, which divides return by the Ulcer Index. The Ulcer Index measures the depth and duration of drawdowns from earlier highs. 

Optimal Portfolios  –  January 2018 – September 2024

Here are some attractive SMURF, BADASS, GLTR,  and QBAT allocations in that order. We can include aggressive assets because of our models’ inherent risk controls and the reduced portfolio risk from using models with modest correlations. 

 
SMURF
QBAT
GLTR
BADASS
S&P500
30/30/25/15
35/30/20/15
35/35/15/15 
CAGR
    23.1
  22.8
 15.4
    57.7
    12.5
       31.8
        32.7
       34.9
STD DEV
    13.7
  16.1
 12.3
    38.4
    18.6
       16.1
        16.2
       17.6
SHARPE
   1.60
  1.37
 1.23
    1.38
    0.72
       1.83
        1.85
       1.81
UPI
   5.12
  4.53
 4.88
    8.67
    1.80
     14.53
      14.47
     14.47
MAX DD
 -12.2
-13.7
 -9.2
  -21.3
  -23.8
        -7.4
        -7.5
        -8.2
AVG DD
   -2.7
  -3.2
 -2.1
    -5.3
    -5.1
        -1.3
         -1.3
        -1.4
W%MOS
     76
    67
   68
       67
      63
          74
            74
           74

Results do not guarantee future success and do not represent returns that any investor attained. You cannot invest directly in our models. CAGR is the compound annual growth rate. Drawdowns are on a month-end basis. UPI is the Ulcer Performance Index which divides return by the Ulcer Index. The Ulcer Index measures the depth and duration of drawdowns from earlier highs. Please see the Disclaimer page for additional information.

Please contact us for more information and fact sheets on all our proprietary models.