AI investment management software

Boosted Insights augments the investment process by empowering investment professionals to source new ideas and create models based on their unique financial domain expertise.

Banner Image

Explore some of the many ways Boosted can empower financial firms and enhance the investment decision process.

Dynamic Stock Screening

A portfolio manager realizes that one of the easiest and earliest ways to implement for machine learning in asset management is to use it to sort equities.

They know they're able to perform many of the same kinds of analysis machine learning can do, but recognize that machines are able to perform calculations at a much faster speed. They also know that some traditional screening tools can be backward looking and do not adapt to new regimes, so they want a stock screen that is constantly adapting and updating itself based on new data.

Boosted Insights takes the PM's capital markets expertise and the variables they find important in stock picking, combines them with big data (either the PM's own or from our many data partners) and produces a backtested model for the PM to analyze in more depth.

The model can be made live, where it will continuously learn from the manager's inputs to predictively rank stocks. It also discovers and explains which combination of features are important in different periods.

A fully ranked stock unvierse with dynamic and adaptive stock rankings. Our finance-specific algorithms compare every stock in a unvierse to every other stock (that's 250,000 comparisons in the S&P 500).

The PM's universe is bucketed into categories based on how our machine learning predicts they will perform (using the inputs the PM found important). Our machine learning has proven excellent at picking the winners from the losers. Sign up for our newsletter below to see for yourself!

Machine Learning Portfolio Overlay

An investment manager is intrigued by the notion of machine learning and would like to start exploring it more. The PM has heard that artificial intelligence might be helpful in either reducing risk or increasing return (or ideally - both).

However, they are concerned that utilizing machine learning may lead to high turnover or unfamiliar names in their book. In order to meet the needs of their portfolio, it is imperative that implementing machine learning in their process must not add any new names and keeps turnover relatively low.

Boosted Insights allows the manager to quickly and efficiently upload their existing portfolio through either CSV or our custom API. Their portfolio is reflected in the platform - complete with tear sheet info and comparisons against their benchmark.

They can use our Boosted Insights' portfolio settings to apply a machine learning overlay to their portfolio, specifying any constraints they might have (low turnover, lower risk, blacklisted equities, and so on).

The machine learning overlay sorts the investment manager's portfolio. In places where the machine learning agreed with the PM's stock picks, it slightly increased its position. Similarly, in places where the machine learning did not agree with the PM's stock picks, it slightly decreased its position. Through these small changes, is able to offer big results.

To read more, check out our white paper here.

Targeted Baskets

A portfolio manager seeks to create a targeted basket of longs or shorts within distinct sectors or with specific attributes.

Typically, they simply short the whole sector or an index of specific funds. They see that using AI and ML can allow them more granularity into their short or long baskets.

The PM applies their stock picking process with Boosted Insights. They select their own variables to create a more targeted basket of stocks.

Boosted Insights is very good at taking the users inputs and separating the winners from the losers, splitting the client's equity universe into quantiles or buckets of outperforming and underperforming stocks.

A highly customized basket of stocks is selected for a specific purpose, allowing the portfolio manager to create additional alpha over using a broad market index.

The PM can use's AI portfolio constraints to meet their exacting standards and ensure that their targeted basket stays within their specific range of needs.

Machine Learning Risk Mitigation

A portfolio manager looks to reduce their risk, both known (factors like momentum, volatility and trading activity) and unknown (including black swan events like Covid-19 or stock events like GME).

They know that being able to time these risk factors will play a critical role in their portfolio management and think that machine learning - which is excellent at seeing patterns that are unintuitive - may be able to give them an edge.’s machine learning can isolate and mitigate risk. The PM can explore risk factors in detail on the app.

Boosted Insights has a detailed factor timing module, where portfolio managers can compare different risk factors against one another, see visual examples in charts, explore stock data and data tables. The factors in the equities within their universe are also graphed by performance. This can help the PM surface when a stock or factor may be due to change or shift.

The multitude of ways to assess and evaluate risk factors allow the portfolio manager to decide how to allocate their portfolio. The Boosted Insights backtested portfolio the PM created can highlight where the PM's investment portfolio is most at risk, along with concrete ways to reduce that risk.

Alternative Data Integration

An insitutional investor believes that their custom data is predictive and wants to generate maximum value from it. They need to be able to run tests on it quickly and efficiently, with flags for problems with their data and normalizations to make their data easy to apply.

They look to be able to upload this data either via API or within an easy to use app and are happy to learn that offers both.

The institutional investor uploads their custom data (alternative data, like credit card, geolocation, insider transactions) to Boosted Insights, either via API or within the app, and creates a model. This data is cross-referenced against their equity universe and any problems with mapping the stocks are flagged for the user.

Boosted Insights determines if the custom data is in fact predictive and generates trade ideas, rankings and a full analysis on the strength of this data. All of this information is surfaced through the simple to use Boosted Insights app and visualized for quick understanding of the value of the alternative data.

The institutional investor can also use functionality to compare portfolios, combine different models and data sets, or highlight which of the features of their alternative data are most predictive and when.

Actively Traded Models

A portfolio manager wants to operate on a fully quantitative basis using machine learning. The manager realizes that building out their own machine learning solution can take years, and even then, there is no guarantee their efforts will work.

They seek to utilize a proven artificial intelligence solution like Boosted Insights, which already accounts for difficulties in using machine learning for asset management, like non-stationarity, various biases that can influence the machine, and high noise-to-signal ratios.

The portfolio manager creates a model (using their own unique variables), sets portfolio constraints and runs back tests on their models. They can perform multiple back tests to see how the model may perform in real time. They can also use portfolio optimzation techniques to further drive performance in their models.

Once the PM is confident of their back tests, they can set the model to “go live” with current data. They then execute the model’s dynamic trading strategy.

A fully quantitative portfolio runs using technical variables and/or fundamental data. The asset manager can market their fully quantitative fund to investors. The manager can also have confidence that their portfolio - though fully quantitative - still combines their capital markets expertise and machine learning power for a quantamental approach.