Dual-Horizon Investment Process
Combine quantitative signals with qualitative analysis of structural change
- Difficulty
- Advanced
- Time to result
- ~ongoing to results
- Steps
- 6
- Confidence
- 95%
The process divides the investment environment by observability and horizon. Short-term drivers that appear at high frequency can be measured and incorporated into quantitative models. Larger structural forces—such as deglobalization, geopolitical fragmentation, demographic change, decarbonization, or digitalization—often lack enough repeated observations for reliable direct modeling. Analysts therefore reason through their causal implications for growth, inflation, market structure, and individual asset classes. The quantitative research process and qualitative overlay are then reconciled into one portfolio view. The overlay is not an excuse for intuition without evidence; it is a governed mechanism for handling changes that are economically important but poorly represented in historical samples.
Origin
Vassalou described the dual approach used by her multi-asset team to incorporate both frequent market data and tectonic changes in the world economy. Extracted from Masters in Business.
Core principles
- 01Observable short-term drivers belong in quantitative models
- 02Structural forces require causal qualitative reasoning
- 03Neither models nor judgment should operate in isolation
- 04The qualitative overlay should address what historical data cannot yet represent
How to run it
- 1
Inventory the drivers
List the developments that could affect the portfolio and distinguish immediate market variables from long-horizon structural forces.
Pro tip Group drivers by both time horizon and data availability.
Watch out Mixing all drivers into one undifferentiated narrative obscures how they should be analyzed.
- 2
Model observable signals
Use quantitative methods for variables that are measurable at sufficient frequency, such as volatility, correlations, prices, or short-term economic releases.
Pro tip Align sampling frequency with the decisions the signal is intended to support.
Watch out High-frequency data can create false precision around noisy relationships.
- 3
Map structural causality
For slow forces, articulate how each could affect growth, inflation, trade, cash flows, and risk premia.
Pro tip Write explicit causal chains instead of relying on labels such as deglobalization.
Watch out A structural theme without a transmission mechanism is not an investment thesis.
- 4
Apply specialist judgment
Use analysts experience in particular asset classes to test whether the proposed effects are plausible and investable.
Pro tip Invite dissent from specialists whose markets contradict the macro thesis.
Watch out Authority or experience should not substitute for a reasoned argument.
- 5
Reconcile the two views
Compare quantitative outputs with the qualitative regime assessment and resolve material conflicts before setting exposure.
Pro tip Record whether the disagreement reflects horizon, data limitations, or a genuinely different forecast.
Watch out An undocumented overlay can become arbitrary discretion.
- 6
Review as evidence arrives
Move structural hypotheses into models when enough observable evidence accumulates, and retire qualitative assumptions that fail.
Pro tip Define indicators that would confirm or falsify each structural thesis.
Watch out Do not preserve a compelling narrative after its predicted effects fail to appear.
In the wild
A multi-asset team cannot estimate a stable deglobalization coefficient from a small number of historical episodes. It instead maps reshoring and trade fragmentation to production costs, inflation, capital spending, and sector opportunities, while quantitative models continue processing current price, volatility, and economic data.
→ The portfolio incorporates a major structural transition without pretending that limited history provides precise statistical estimates.
Common mistakes
Forcing structural change into weak data
A model may assign spurious precision to a development with too few comparable historical observations.
Using an ungoverned overlay
Qualitative judgment should have explicit reasoning, dissent, and falsification criteria rather than functioning as a discretionary override.
Ignoring horizon differences
A short-term model and a structural thesis may appear contradictory while actually addressing different time periods.
Is it for you?
Best for
It is best for multi-asset teams investing through deglobalization, policy shifts, geopolitical change, and other regime transitions.
Not ideal for
It is not ideal when decisions are purely mechanical or when the team cannot articulate and govern its qualitative judgments.
From the transcript
“We have a dual approach, so we certainly have our research process that it's based on models that we have created and we keep evolving.”
“The hardest part is to incorporate the bigger picture, and that's really where the qualitative overlay comes into play.”
From the episode
Maria Vassalou on the Small-Cap Effect
Maria Vassalou