Turing Technology White Paper:

January, 2024
Ensemble Active Management: AI's Transformation of Active Management
Authors: Vadim Fishman, Rober Nestor

This pioneering White Paper is the largest study ever conducted on Ensemble Active Management (EAM), based on 60,000 portfolios and over 560 million data points. The Paper details the two-levels of advanced technology used to construct EAM portfolios, as well as the cause-and-effect of how these Machine Learning tools are able to generate "Ensemble Alpha", a new form of investment value-add.

Key takeaways include that 1) EAM portfolios were able to achieve rolling period performance success rates of 74% to 85% vs traditional benchmarks, and delivered 2) average annual excess returns of between 450 to 500+ basis points (4.5 to 5+%).

Key Topics Examined:

  • EAM design and construction, including the methodology for extracting the 'predictive engine' from mutual funds.
  • Use of Ensemble Methods to create a new source of alpha. Ensemble Methods is the 'go-to' Machine Learning technology used in more than 250,000 published applications to solve the world's most difficult computational challenges.
  • Relative performance and success rates versus industry benchmarks and traditional actively managed mutual funds.

Key excerpts:

  • "EAM succeeds by applying the proven mathematics of Ensemble Methods, which are at the heart of nearly every major computational challenge in the world, to a multi-investment-manager foundation."
  • "The results are profound, and statistically significant [with] EAM systematically outperforming both benchmarks and the underlying portfolio of funds."
  • "EAM portfolios were able to reduce tail risk vs mutual funds."

Turing Technology Research Paper:

October 3, 2019
The Singular Impact of High Conviction Overweight Positions for Active Managers
Author: Alexey Panchekha

This groundbreaking research reflects the first set of findings to emerge from Turing's Hercules Database. This database contains up to a decade's worth of real-time, daily holdings and portfolio weights of actively managed mutual funds from more than $2.25 trillion in fund assets under management, reflecting data covering 70+ fund families and hundreds of funds. This database allows Turing to see into active funds at the 'DNA' level.

Key Topics Addressed:

  • Manager Skill: Active managers deliver significant stock selection skill, but it is limited to only one set of stock decisions;
  • Active Manager Paradox: Research identified a 'genetic defect' that is impairing actively managed funds from succeeding versus index funds and ETFs;
  • Path Forward: Research also identified insights into approaches for active managers that can 'cure' their structural flaw, creating higher statistical expectations for outperformance.

Turing Technology Research Paper:

January 5, 2019
Understanding Ensemble Active Management - Innovation in Action
Authors: Alexey Panchekha, CFA, Matthew Bell, CFA, and Robert Tull

EXCERPT: "The key takeaway should be that the investment industry needs to treat the construction of the Alpha Engine and the Beta Anchor as two discrete activities, and the portfolio design function should have as an explicit goal the minimization of the Beta Anchor and its dilutive impairment on performance.
The new insight is that the requirement for the Beta Anchor can be reduced, or even eliminated, if a second layer of diversification at the investment strategy level can be introduced. The simplistic beauty of EAM is that it uses a new layer of diversification to de-risk the Alpha Engine before pairing it with the Beta Anchor. Therefore the Beta Anchor's role is automatically lessened, its scale can be reduced, and the natural benefits the Alpha Engine can be delivered directly to the investor."

Turing Technology White Paper:

September 5, 2018
ENSEMBLE ACTIVE MANAGEMENT: The Next Evolution In Active Investment Management
Author: The EAM Research Consortium; Contributing Editors: Alexey Panchekha, CFA, Matthew Bell, CFA, and Robert Tull

EXCERPT: "This White Paper questions the superiority of the traditional Active Management paradigm. Do stand-alone, 'single-expert' investment managers or management teams, with well-defined yet rigidly entrenched philosophies and methodologies, deliver optimal results? The conclusion, derived from a database reflecting 30,000 test portfolios and 165 million data points, was that they do not.
A new approach to investment management, referred to as "Ensemble Active Management" and representing the intersection of Artificial Intelligence and traditional Active Management, was proven the superior option."


Institutional Investor, by Julie Segal
The 'Genetic Defect' in Active Management

EXCERPT: "Active managers can produce persistent excess returns when it comes to their best ideas — but they only devote half of their capital to these investments. The result has been a shredding of the industry as investors have moved to passive index funds, according to new CFA Institute research.
Active managers have historically created highly diversified portfolios as a risk management tactic. But the approach is putting active managers permanently in the penalty box."

CFA Institute: Enterprising Investor, by Alexey Panchekha, CFA; Edited by the CFA Institute
The Active Manager Paradox: High Conviction Overweight Positions

EXCERPT: "While our data shows that fund managers can exhibit persistent skill through their high-conviction best ideas, it also reveals a portfolio design paradox.
As the sole source of excess return, High-Conviction Overweights need to be the main emphasis of all actively managed portfolios. Any allocation to anything else will reduce returns."

CFA Institute: Enterprising Investor, by Alexey Panchekha, CFA, Matthew Bell, CFA, and Robert Tull; edited by CFA Institute
Ensemble Active Management (EAM): Taming Toxic Tails

EXCERPT: "If EAM portfolios can deliver on their early promise, then Ensemble Active Management may prove to be the disruptive innovation that the active management industry has been searching for and thus could redefine the competitive balance between active and passive management."

Institutional Investor, by Julie Segal
Diversification is Causing Active Managers to Underperform. AI Could Fix It.

SUB-TITLE: "Proven artificial intelligence and machine learning techniques could help protect active managers' best ideas by offering a risk management alternative to diversification."

Institutional Investor, by Julie Segal
Is This the Silver Bullet That Will Save Active Management?

EXCERPT: "Active managers under threat from the boom in passive investment funds can be saved from oblivion by using artificial intelligence techniques long used successfully to power everything from Netflix to forecasting hurricanes, argues a new consortium of academics, data scientists, and technology and investment professionals."



Goldsman, D., 2024, Methodology, Design and Data Integrity Validation Study of Turing's Ensemble Active Management, Georgia Institute of Technology, Atlanta, GA

This is an independent report written after a months' long academic evaluation of the 2024 EAM White Paper "Ensemble Active Management: AI's Transformation of Active Management." Professor Goldsman and his team were granted complete transparency to the White Paper's methodology, stated biases, input and output data, and even code. They rebuilt the key results from the White Paper using alternative analytical and sampling techniques, validating the published results of the White Paper.


  • Core research methodology.
  • Fund selection and portfolio of funds construction methodology.
  • EAM construction methodology.
  • Performance calculations and assessment of performance outcomes.


  • "Across all portfolio fund style boxes, the EAM portfolio has an overall expected performance benefit of 400-500 basis points when compared against the corresponding [ ] benchmark."
  • "Turing's claims that EAM performance is comparatively better than traditional active management and standard industry benchmarks were also substantiated."

Pinsky, E., 2018, Mathematical Foundation for Ensemble Machine Learning and Ensemble Portfolio Analysis, Boston University, Boston, MA.
EXCERPT: "In many areas, combination of models often perform better than individual models. This paper focuses on application of this approach to construct large-cap portfolios from individual large-cap mutual funds."