We in the Jupiter systematic team adopt a highly active approach. However, our approach is very different from traditional active management, often known as discretionary management. Rather than employing traditional techniques, such as manually scrutinising company annual reports, meeting management teams, and studying third-party analysis by hand, we use computer-based techniques to analyse huge volumes of publicly available information. This allows us to scrutinise a large universe of global stocks against a number of proprietary stock selection criteria, which we have developed and refined over years. Although our process is systematic, the principles on which our stock selection criteria are based are very intuitive and easily understood by investors. They include well-known principles such as share price value, balance sheet quality, company growth characteristics, efficient use of capital, analyst sentiment and supportive market trends.
Our investment process is data-led, rather than opinion-led
We apply a disciplined, scientific methodology to testing hypotheses. We believe that systematic investing has an advantage because it is dispassionate rather than, as so much investment is, prone to human bias. Indeed, our process seeks to exploit other investors’ psychological and behavioural biases. We do not believe that markets are efficient, and we seek to exploit the inefficiencies we find. We seek to fine-tune and improve our investment process iteratively in a rigorous way over time.
Over recent years we have introduced numerous improvements and enhancements to our investment process. For example, we have broadened the data we analyse. This allows us to achieve greater diversification, we believe. New kinds of information we have incorporated over the last few years include ESG (Environmental, Social & Governance) data, company executive transactions (from regulatory filings), management sentiment (from textual analysis of company earnings calls) and buy-side sentiment (from trade and fund flows data). The diagram shows some of the main enhancements we have introduced to our investment process over the last few years.
In September 2019, we introduced value quality decoupling, which allows greater flexibility within our dynamic valuation stock selection strategy. In total, our investment process includes five distinct and diversified stock selection criteria. One of these, which we call ‘dynamic valuation’, seeks to identify shares with attractive valuations. It does this by analysing company data in a similar way to value investors, except that by being systematic it is able to accomplish this at a large scale, analysing the financial metrics of thousands of companies.
Our dynamic valuation strategy is emphatically not a generic value factor: it is proprietary. It differs from generic value factors in numerous ways: for example, it also incorporates analysis of the quality of companies, because our studies indicate this can reduce the downside risk of value investing. The introduction of value quality decoupling in 2019 additionally meant we have been better able to navigate periods where both value and quality styles are out of favour. It therefore gives our investment process more flexibility.
Our allocation to the five stock selection criteria is not static, but dynamic – it changes over time, responding to our measurements of the market environment. Also in September 2019, we enhanced our consideration of conditional downside risk, designed to improve how we weigh the risk of factors in different types of market environment. Both our value quality decoupling enhancement and our conditional downside risk improvement allow our dynamic weighting scheme to operate with greater flexibility and responsiveness.
In September 2019 and January 2020, we made two important enhancements to our portfolio construction and risk management framework. In September 2019, we introduced an additional statistical risk model. The Principal Component Analysis (PCA) model bolsters our existing factor-based risk model framework and helps us to capture time-varying risks and understand large, complex data sets with many dimensions. The new model can automatically identify and control transitory sources of risk without the need to prespecify them, such as those driven by geopolitics and macroeconomics. In January 2020, we introduced revised constraints and enhancements to how country, sector and industry effects are controlled at the factor design stage, as well as at the portfolio construction stage, with the aim of improving risk-adjusted returns.
While the latter two enhancements were designed to improve portfolio construction, the next three were to improve the way in which we select stocks. In April 2020, we added a new component to our company management stock selection strategy, to extract information from directors’ trades in their own company shares. In June 2020, we incorporated environmental, social and governance (ESG) metrics. We look not only at current levels but also at changes in them. One of the problems faced by ESG investing is that environmental, social and governance factors may in part overlap with other factors (for example, with company quality). We therefore took care to ensure that by increasing our exposure to E, S and G factors we were able to avoid accidental, unintended style tilts.
In November 2020, we added a new component to our analyst sentiment strategy to capture management sentiment and quality signals from transcripts of management earnings calls. This uses Natural Language Processing (NLP) to extract useful information from large bodies of text.
In March 2021, we made another enhancement to our dynamic weighting scheme. We called this ‘high conviction rotations’, and it was another step in improving flexibility and responsiveness. It helps us better identify relationships between market environment indicators and factor return expectations and allows larger rotations between factors when the model has higher conviction.
In November 2021, we introduced a new component to extract useful information from fund flows into different types of equity funds. Fund flows are the cash that goes in and out of mutual funds, as investors decide how to allocate or reallocate their money. A problem with this kind of data in the past has been its low frequency and lack of timeliness – it has not been very up to date. We were able to source data that was frequent and timely enough for us to develop useful signals.
In December 2022, the reputational risk enhancement was introduced. This involved the incorporation of ESG-related reputational risk based on news items. We determined that this was helpful in allowing us to identify stocks with returns being driven by characteristics other than traditional factors.
In October 2023, we introduced a new signal to benefit from global industry equity fund flows. For more information on global industry flows, see here.
We introduced an expansion of our trading universe by about 500 stocks in October 2023. This is not an addition to our portfolios, of course, but an increase in the number of stocks we analyse daily and consider for inclusion in portfolios, long or short. The expansion allows us to benefit from market liquidity, and the greater breadth gives a greater alpha opportunity.
In November 2024, an evolution in the model’s company management stock selection strategy was the addition of patent data. Patents give the legal right to use an invention. Our patents signal enables us to identify companies which operate with high innovation efficiency based on their ability to translate a project into potential future earnings, through the use of patents.
In March 2025, we improved our dynamic weighting scheme by introducing a dynamic valuation enhancement to better navigate more extreme market environments when investors are focusing on either deep value or expensive quality stocks.
We further enhanced our sentiment stock selection strategy in September 2025 by refining the construction of one of its key sub-components. The updated methodology more effectively identifies short-term dislocations and, importantly, improves our ability to distinguish between price dislocations that are likely temporary and those that may prove permanent.
Shortly afterwards, in December 2025, we improved our portfolio construction and risk management framework by an enhancement to volume forecasting. This improves the capture of volume fragmentation across trading venues.
In February 2026, we made another improvement to portfolio construction, increasing the efficiency of alpha implementation by optimising the sizing of trades. We call this our trade thresholds enhancement.
In March 2026, we introduced another extension to our company management stock selection strategy, by capturing the non-linear impact of management decisions on price action trends. We call this latest enhancement management signalling, and further detail about it is on our website.
We continue to pursue a highly active research programme in house. We also continue to work with a number of academics on promising emerging areas of research. The enhancements we have introduced in recent years are designed to increase diversification, use new sources of information, and allow greater flexibility, all with the aim of continuing to improve risk-adjusted outcomes.
Important information
Marketing communication.
This document is intended for investment professionals* and is not for the use or benefit of other persons, including retail investors. The value of investments and income may go down as well as up and investors may not get back amounts originally invested. Exchange rate changes may cause the value of investments to fall as well as rise. This document is information only and is not investment advice. The views expressed are those of the author(s) at the time of preparation, are not necessarily those of Jupiter as a whole and may be subject to change. Every effort is made to ensure the accuracy of any information provided but no assurances or warranties are given. Past performance does not predict future returns. Forecasts are an estimate of future performance based on evidence from the past on how the value of this investment varies, and/or current market conditions. Forecasts are not a reliable prediction of future returns. What you will get will vary depending on how the market performs and how long you keep the investment/product. Historical examples are for illustrative purposes only, may not represent current fund holdings, and are not a recommendation to buy or sell. Where a benchmark is used for comparison, it is shown for illustrative purposes only and does not imply future performance.No part of this document may be reproduced in any manner without the prior permission of Jupiter. Issued in the UK by Jupiter Asset Management Limited (JAM), registered address: The Zig Zag Building, 70 Victoria Street, London, SW1E 6SQ is authorised and regulated by the Financial Conduct Authority. Issued in the EU by Jupiter Asset Management International S.A. (JAMI), registered address: 5, Rue Heienhaff, Senningerberg L-1736, Luxembourg which is authorised and regulated by the Commission de Surveillance du Secteur Financier. No part of this document may be reproduced in any manner without the prior permission of JAM/JAMI.
*In Singapore, investment professionals means accredited and/or institutional investors as defined under section 4A of the securities and futures act (“SFA”) and in Hong Kong professional investors as defined under the Securities and Futures Ordinance (Cap. 571 of the Laws of Hong Kong).