Insighful and rigorous data analysis drives good investment decision making.


We exploit big data and machine learning concepts in constructing proprietary, data-driven algorithms to reveal investment opportunities.

We don't attempt to forecast the overall market or the macro economy. Instead, we seek out situations where we anticipate upward movement in promising individual stocks.

We have more confidence in short term forecasts than in predictions of far-off events, so we prefer a relatively short investment horizon, typically holding positions for about a month.

Our approach is empirical but exploits well documented investment principles like value and momentum:

  • Side step the bubbles. We screen the universe of US large cap equities for well priced stocks, rejecting stocks overpriced compared to their fundamental value. We don't overpay.
  • Go where the models work best. Our statistical models work better on undervalued stocks than for the market as a whole. That's where we focus.
  • Understand the flow of the markets. We use momentum as a gauge of market sentiment. We attempt to buy on the dip, but only when we have seen a pattern of stocks recovering from prior dips.
  • Spread the risk. We diversify our portfolios to protect clients from excessive volatility.

We continuously assess our parameters and strategies, and are always researching new ideas.