Raltin - Financial Data Analytics & Research

How Does Raltin’s MLF Data Feed Help Investors?

Many institutional investors spend an inordinate time developing features to feed into their machine learning models. As Andrew Ng, the co-founder of Google Brain, said - “Coming up with features is difficult, time-consuming, requires expert knowledge. "Applied machine learning" is basically feature engineering.”

Raltin’s cross functional team of software engineers, data scientists, and financial domain experts have developed the most comprehensive set of features for your machine learning models.

Raltin’s MLF data feed offers three primary advantages to institutional investors:

• Features that cannot be developed in house: Some features such as correlations across the entire equity and options universe require running billions of correlations and are not possible to run in house because of computing costs and engineering resources.
• Raltin’s proprietary features: Raltin has invested years in developing proprietary signals that give investors an edge. An example of a propieratory feature includes forecasts that give probability estimates on future volatility of equities.
• Cheaper than attempting to do it in house: The sheer number of features available in Raltin’s MLF data feed would make it cost-prohibitive to replicate it in house.

What kind of features are included in Raltin’s MLF Data Feed?

Raltin’s MLF Data feed provides end of day updates across hundreds of features including:

• Correlations across the entire US Equity universe: Raltin runs billions of correlations on price, implied volatility (IV) and other metrics across the equity and options universe to show the most (and inversely) correlated stocks/ETFs for any given ticker.
• Raltin’s proprietary features: Raltin integrates price/volume, options, fundamentals, and other Raltin proprietary data to help investors give an edge. An example of a Raltin proprietary feature is short term volatility forecast on the underlying equity.
• Fundamentals: Fundamentals such as net cash are juxtaposed with other data sets such as equity options data to come up advanced metrics such as “Net Cash - % Distance from current stock price”.
• Technical Indicators: These technical indicators include simple features like moving averages and more advanced indicators such as Chande Momentum Oscillator (CMO).
• And many many more coming: Raltin is supported by a team of 40+ engineers, data analysts, statisticians and domain experts that are continuously adding new features to the MLF data feed.