Proprietary Models

We designed our public dual momentum model, GEM, to be simple and easy to use by do-it-yourself investors. GEM introduced dual momentum to the world.

The aim of our proprietary models is to achieve the most consistent returns possible using all available investment tools. We license our proprietary model signals to substantial private investors and select investment advisors.

Our proprietary models strive to be highly adaptive to market conditions and can serve as the basis for core portfolios. They have their foundation in academic research supported by out-of-sample and real-time validation. These are our proprietary models:

Dual Momentum Fixed Income (DMFI)
Dual momentum works as well with fixed income as it does with equities. DMFI applies dual momentum to the short and intermediate segments of the bond market. Since 1970, DMFI has had the same return as the S&P 500 but with considerably less volatility. DMFI is an attractive substitute for passive bonds in a stock-bond portfolio.
Enhanced Global Balanced Momentum (E-GBM)

E-GBM is a balanced allocation between stocks, bonds, and other assets. It is a general-purpose dual momentum model suitable for most investors. 

Advanced Global Equities Momentum (A-GEM)

A-GEM focuses on the stock market but includes bonds and other assets when market conditions warrant it. A-GEM adds intermarket relationships to dual momentum for determining trends. A-GEM uses daily and weekly data.

NASDAQ Breadth and Trend (QBAT)

QBAT applies breadth, trend, and mean reversion to the ProShares Ultra QQQ ETF (QLD). When not invested in QLD, QBAT holds the same positions as A-GEM. QBAT uses only daily data. It is our most responsive model.

Performance

Here is the performance of our Dual Momentum Fixed Income (DMFI) model compared to Aggregate Bonds, 10-Year Treasury Bonds, and the S&P 500 index.

DMFI Model Performance January 1970 through March 2023 

     DMFI   AGG   10 YR   S&P 500  Balanced
CAGR     10.7      6.9       7.2     10.5         9.5
Standard Deviation       5.1      5.3       8.0     15.4      10.2
Sharpe Ratio     1.14    0.40    0.34     0.43      0.50
Ulcer Index    1.33    2.47    3.99  12.88      5.90
Worst Drawdown    -6.1  -17.2  -21.0  -51.0    -29.7
Avg Drawdown    -0.6    -1.2    -2.3    -7.5     -3.2
% Up Months      82      68      60      63       64

Results do not guarantee future success and do not represent returns that any investor attained. You cannot invest directly in our models. AGG is the Bloomberg Barclays U.S. Aggregate Bond Index from its starting date of July 1983 and the Ibbotson 5-Year U.S. Government Bond Index before that. 10 YR is the ICE U.S. Treasury 7-10-Year Bond Index. Balanced is 60% S&P 500 Index and 40% ICE U.S. Treasury 7-10-Year Bond Index. CAGR is the compound annual growth rate. Drawdowns are on a month-end basis. Please see the Disclaimer page for additional information.

                                                                  Dual Momentum Fixed Income January 1970 – March 2023

Here are the back-tested results of E-GBM compared to the S&P 500, a 60/40% balanced stock/bond portfolio, and our public GEM model.

E-GBM Model Performance January 1970 through March 2023

 

   E-GBM

    GEM

 S&P 500

   Balanced

CAGR

       17.3

      15.2

      10.6

        9.3

Standard Deviation

        9.7

      12.6

      15.4

      10.1

Sharpe Ratio

      1.27

      0.87

     0.46

     0.52

Worst Drawdown

      -8.7

    -19.6

   -51.0

    -29.7

Average Drawdown

      -1.2

     -3.1

     -7.6

     -3.3

% Up Months

        73

       68

       63

       63

Historical data and analysis should not be taken as an indication or guarantee of future performance. We have subtracted transaction slippage of 0.1% per trade. Performance does not represent actual fund or portfolio performance. Performance includes reinvestment of interest and dividends. CAGR is the compound annual growth rate. Balanced is 60% S&P 500 Index and 40% ICE U.S. Treasury 7-10 Year Bond Index. Worst drawdown is on a cumulative month-end basis. Please see the Disclaimer page for more information. 

Here are the results of the current versions of our A-GEM and Enhanced Global Balanced Momentum (E-GBM) models compared to the S&P 500, a 70%/30% balanced stock/bond portfolio, our public Global Equities Momentum (GEM), and a 50/50% blend of A-GEM/E-GBM. Results begin in January 1999 due to limited daily data before then. We have extended ETF data with associated index data when necessary.

A-GEM Model Performance January 1999 through March 2023

 S&P 500      Balanced   GEM E-GBM A-GEM 50/50
CAGR      7.0           6.5      9.5  13.2    17.3   15.3
STD DEV   16.8        11.5   11.9    9.5    11.0    9.3
SHARPE   0.48        0.60   0.82  1.35    1.51  1.58
MAX DD -52.9      -36.6  -19.6  -8.3  -11.3  -8.3
AVG DD -10.7          -5.3    -4.5  -1.5   -1.5  -1.2
% UP MOS      61          64      67    69    70    70

 

Results do not guarantee future success and do not represent returns that any investor attained. We have subtracted estimated transaction slippage of 0.1% per trade. Performance includes reinvestment of interest and dividends. CAGR is the compound annual growth rate. BALANCED is 70% S&P 500 Index and 30% ICE U.S. Treasury 7-10 Year Bond Index. Maximum drawdown is on a cumulative month-end basis. See our Disclaimer page for more information.

Here are the back-tested results of QBAT compared to QQQ, the S&P 500, Global Equities Momentum (GEM), Enhanced Global Balanced Momentum (E-GBM), and Advanced Global Equities (A-GEM). We also show 50%/30%/20%, and 50%/40%/10% allocations to A-GEM, E-GBM, and QBAT respectively, as well as a 25% allocation to each plus Dual Momentum Fixed Income (DMFI). The 50%/50% allocation is E-GBM/A-GEM. 

QBAT Model Performance July 2006 through March 2023

 QQQE-GBMA-GEMQBAT  25 ALL50/30/2050/40/1050/50
CAGR  14.3 11.2   15.3  39.6     19.1    18.9  16.1 13.3
STD DEV  19.6  8.6   11.1  23.7     10.3    11.5 10.2   9.1
SHARPE  0.781.28  1.35  1.55     1.76    1.57 1.52 1.42
ULCER13.692.47  3.12 4.62    2.20   2.78 2.56 2.41
MAX DD-50.2-7.3 -11.3-18.6    -9.7 -10.5 -9.4  -8.3
AVG DD  -7.2-1.5  -1.7  -2.5     -1.1   -1.4  -1.4  -1.3
W% MOS    66  67    68    66       70     68   69    68

3

Portfolio Effect

Our models do best when used in combination because of their strategy differences and modest correlations. Here are the monthly correlations from July 2006.

   QBAT  A-GEM  E-GBM  DMFIS&P 500QQQ
 QBAT     1.00     0.73    0.58    0.26   0.50   0.58
 A-GEM      1.00    0.71    0.42   0.51   0.51
 E-GBM      1.00   0.48   0.52   0.49
 DMFI      1.00   0.22   0.18
 S&P 500       1.00   0.92

We can also combine QBAT with DMFI to create a “barbell approach” of non-correlated low and high-volatility strategies. Here are the back-tested results from July 2006 with 50%/50%, 40%/60%, 30%/70%, 20%/80%, and 10%/90% in QBAT and DMFI, respectively. Since QBAT is invested the same as A-GEM when it is not in QLD, these allocations are an amalgamation of three different models. Even a 10% allocation to QBAT considerably improves portfolio results while preserving a low risk profile.

QBAT with DMFI Model Performance July 2006 through March 2023

 

QQQ

DMFI

QBAT

 50/50

 40/60

 30/70

 20/80

 10/90

CAGR

 14.3

 10.1

  39.6

  24.9

   22.0

   19.0

  16.0

  13.0

STD DEV

 19.6

  4.9

  23.7

  12.7

  10.7

   8.7

   6.9

   5.6

SHARPE

 0.78

 1.97

  1.55

  1.82

  1.93

  2.05

 2.19

 2.24

MAX DD

-50.2

 -5.3

-18.6

-11.1

  -9.6

  -8.1

 -6.6

  -5.1

AVG DD

 -7.2

 -0.5

  -2.5

  -1.1

 -0.9

 -0.7

-0.6

 -0.5

W% MOS

   66

   80

    66

    71

   73

    74

   78

   82

Results are not a guarantee of future success and do not represent returns that any investor actually attained. Performance includes reinvestment of interest and dividends. CAGR is the compound annual growth rate. Maximum drawdown is on a cumulative month-end basis. Leveraged ETFs have high risk and extreme volatility. Please see our Disclaimer page for more information.

Contact us for fact sheets and additional information about our Proprietary Models.