The FUTURE OF FINANCE

 
 
 

We are living at a historic moment of technological change. For the first time, powerful supercomputers permit the application of Machine Learning (ML) techniques that can extract insightful information from complex financial datasets.


WHY QUANT INVESTMENTS OFTEN FAIL

For the past 60 years, most quantitative asset managers have applied econometric models, with disappointing results:

  • The quantitative funds offered by most asset managers are based on false discoveries, as a result of backtest overfitting.

  • The complexity of financial systems overly exceeds the modeling capability of Econometrics.

  • Some of the most interesting datasets, such as satellite images, consumer transactions, or news articles, are beyond the grasp of these traditional quantitative methods.

Instead of Econometrics, the most successful hedge funds in history use ML as their primary quantitative tool, because of ML’s flexibility and ability to identify complex patterns with minimum assumptions.

Although ML is extremely effective, its correct application to financial time series is not straightforward, requiring domain-specific knowledge and experience. Predicting financial outcomes is orders-of-magnitude more difficult than, for example, recognizing faces or driving cars. When applied incorrectly, ML compounds the risk of backtest overfitting.

As a result of backtest overfitting, the quantitative funds offered by most asset managers are based on false discoveries. ML can prevent backtest overfitting by accounting for selection bias.

THE TRUE POSITIVE ADVANTAGE

1_dsr.png

ML is modern statistical analysis. As it is common with any breakthrough, there is much confusion about what ML can do in finance, and how it should be used. The answer is not as simple as “let’s hire some PhDs from Silicon Valley.”

To prevent backtest overfitting, firms must establish robust scientific protocols that control and account for the probability of false investment strategies (false positives).

Unlike most firms that claim to have expertise in ML, True Positive Technologies (TPT) has a certified track record at managing multibillion-dollar mandates for institutional investors. We take pride at helping clients solve challenges that we have previously faced and overcome.

Technologies developed by TPT’s team have been sought after by some of the most successful hedge funds. Leading asset managers have tested our techniques live, and purchased particular implementations of our IP (including several patents) in multimillion-dollar deals. TPT helps modernize institutional investors, by deploying these tried-and-true solutions to their particular investment problems.

TPT has a certified track record at managing multibillion-dollar mandates for institutional investors.