Successful investing is ultimately based on accurate forecasting.
Here's a brief description of our methodology:
Compared to new quantitative data, new qualitative data diffuses more slowly across market participants, and is only gradually incorporated into asset prices. This time delay, plus the accuracy of their mental models, enables investors to avoid large losses by taking action before the market's dominant narrative changes.
We focus on providing insights about the evolving dynamics of the global macro system and the emergent threats they produce that lie beyond the detection horizon and analytical capabilities of quantitative algorithmic methods. We then translate these into probability forecasts for different macro regimes, and use them to adjust our model portfolio's asset allocation. That's our edge — and potentially yours too, if you subscribe.
Specifically, using a method called Multipath Analysis we collect and synthesize high value information (threat indicators and surprises) in the areas of technology, health and disease, energy and the environment, the economy, national security, society, and politics. Their complex interaction over time produces the effects we later observe in the form of investor beliefs and behavior, from which emerge financial market valuations, volumes, and returns.
Rather than statistical and machine learning, our approach more closely resembles estimative intelligence analysis, which employs a combination of bottom-up and top-down sensemaking processes. These were described in Pirolli and Card's classic article, "The Sensemaking Process and Leverage Points for Analyst Technology”, which includes this detailed graphic:
Our forecasting process also draws on lessons Tom Coyne learned from spending four years as a member of the Good Judgment Project team, which won the Intelligence Advanced Research Projects Activity’s forecasting tournament with results that were more than 50% more accurate than the tournament's control group of professional intelligence analysts (the team's experience is described in Professor Philip Tetlock's book, “Superforecasting").
For example, we pay careful attention to base rates, and take a Bayesian approach to using new evidence to update our prior regime probabilities.
