Tenure Maker: An Interactive Example

What is tenure making? And why should I care?

Graphic on left illustrates the behavior of the trading strategy against the market behavior, by measuring the Sharpe Ratio (SR). The green line represents the trading strategy behavior. The blue line represents the market behavior. The purpose of this online interface is to show how the green line gets more and more profitable over time as the program continues to optimize the system to fit historical data. In a matter of seconds or minutes, the program creates what appears to be a very profitable equity curve (with a very high Sharpe Ratio (SR)) based on the input historical stock price dataset. Graphic on right shows inflation in the Sharpe Ratio (SR). As you may see SR values are inflated; we are predicting future realizations of the series by using past realizations, which is of course impossible by construction (the series is IID - independent, identically distributed).

A short introduction:

There are two broad types of investment strategies:
1) Those based on general trading rules. Examples are many trading indicators and seasonal opportunities.
2) Those based on forecasting equations, like econometric models.

Our (Backtest Overfitting Demonstration Tool or BODT) is an example of the first type. For the second type, we have developed the Tenure Maker Simulation Tool (TMST). We call it the tenure maker because many academic research results in finance are subject to the criticism that they have been produced in this way.

  
By checking this box, you acknowledge that this web site is a demonstration of a mathematical concept known as the "tenure maker". The software and material on this site should NOT be interpreted as a recommendation to buy or sell any security or securities; as applicable to the specific investment needs of any particular individual or organization; or as applicable to the forecast of future market prices or trends.
    

Option 1: Full (generates all strategies, no computational time limitation)

  
Please be patient! The program may take up to two minutes to display the results page after clicking Go! button
    

Option 2: Limited (limited by the maximum allowed computational time)

Maximum Computational Time An integer; ranges from 30 to 900. The total computational time in seconds allowed for the program to generate strategies.
Please be patient! The program may take up to two minutes in addition to the entered 'maximum computational time' to display the result page after clicking Go! button

Default values: If you do not enter a value or enter a value that is outside the range mentioned above, a default value will be used. The default value for the maximum computational time is 90 seconds.

About execution time: The number of the strategies generated significantly affect the number of iterations performed by the program.

A more detailed explanation of options is provided in the tutorial.


How Does It Work?

This online tool is for investment strategies that are based on forecasting equations, like econometric models. It
  1. Generates series of IID (independent, identically distributed) Normal returns.
  2. Generates automatically a large number of time series models, where the series is forecasted as a fraction of past realizations of that same series, including transformation such as:
  3. Applies a forward-selection algorithm on these alternative specifications.
  4. As the selection algorithm progresses, it publishes the improved model.

A few recent articles explaining the backtest overfitting in the financial press:

Related documents:

Tutorial of the online tool:

Questions or comments:

Please send any comments or questions for this site to: Amir Salehipour.

Credits:

The project team includes David H. Bailey, Jonathan M. Borwein, Marcos Lopez de Prado, and Amir Salehipour. This program and web page were constructed by Marcos Lopez de Prado and Amir Salehipour.

We gratefully acknowledge the helpful comments and suggestions from colleagues and friends in shaping this web site. Their suggestions were extensive and very helpful in improving readability of the web pages. Thanks!