MSCI has launched SignalLab to give investment teams a common environment for comparing and deploying investment signals. Its 7 October announcement brings Risk Premia and Micro-Industries together with existing FactorLab and Crowding datasets, with more than 600 signals across the platform. Customers can also introduce proprietary data.[1]
The intended operational benefit is reduced data preparation. Signals assembled from different providers can use incompatible identifiers, universes and definitions. MSCI says its library standardises and documents those inputs. Its product page offers access through an API and Snowflake, allowing teams to connect the data to their own research processes.[1][2]
The platform-wide count should be read alongside each module’s specification. The current Risk Premia factsheet, checked on 21 October, gives a narrower picture of that particular library.[3]
Risk Premia module: published scope
| Item | Published specification |
|---|---|
| Signal count | 45 unique signals; 86 daily and monthly variants |
| Equity coverage | US available; global equities scheduled by 30 October 2026 |
| History | Varies by descriptor; earliest starts in 1992 |
| Governance information | Definitions, coverage, update frequency, quality checks and data lineage |
MSCI also published research that helps frame how customers might evaluate the inputs. In a historical study of 41 daily signals, using US equities and data through May 2026, the mean information ratio fell from 1.00 before publication to 0.28 afterwards when eight style-factor exposures were neutralised. That is a risk-adjusted performance statistic, not a percentage return.[4]
The same study illustrates why implementation costs matter. A simulated market-neutral portfolio had a gross information ratio of 1.7 but only 2.4 basis points of round-trip break-even transaction costs. Adding a 10% daily turnover limit reduced the ratio to 1.6 while raising the break-even cost to 6.2 basis points. These are modelled results, not live client returns.[4]
For a trading team, the implication is practical: adding more candidate signals can increase the demand for portfolio construction and execution discipline. Coverage must fit the traded universe, the historical test must use information available at the time, and the expected return must survive the firm’s actual trading costs. A ready-made dataset can shorten preparation while leaving those decisions with the manager.
MSCI’s launch also places researcher validation after AI-assisted production of Risk Premia signals and model-team validation after its Micro-Industries classifications. Human review is therefore part of the published production process, rather than an inference that every machine-generated output is released directly.[1]
Sources & further reading
- MSCI launches SignalLab giving investment teams access to trusted and integrated signals MSCI · 7 Oct 2026 · Accessed 21 Oct 2026
- SignalLab — workflow access and research tools MSCI · Accessed 21 Oct 2026
- MSCI SignalLab Risk Premia module factsheet MSCI · Accessed 21 Oct 2026
- Constructing portfolios from published equity research MSCI · 7 Oct 2026 · Accessed 21 Oct 2026