(951) 268-7836 info@authintel.com

Our Trading Optimizer (TM) technology empowers customers using a rigorous, scientific end-to-end investment management platform that consistently provides higher returns on investments.

Our intelligent automated system offers state-of-the-art data analysis that spawns sophisticated R&D techniques and automatically improves and advances investment strategies based on rapidly changing markets.

We maximize the value of human learning in a software approach by incorporating patent-pending software that consistently evaluates its own performance, modifying its own behavior to adapt to today’s rapidly changing markets.

Focuses on intelligent decision making and prediction accuracy of longer-term trends rather than functioning as a high-frequency or day trading system.

Quick Market Overview
Investment Management Firms Today:

  • Focus on medium-to-long-term profitability by providing investable funds, ranging greatly in size and purpose.
  • Some focus mainly on hedging while others are oriented toward general long-term profitability.
  • May host their own databases or utilize outside services for data.
  • Often utilize analytical packages and professional trading products, but their focus typically is more fundamental than technical.
  • Need tools to help them study trends, current events, and apply analytical methods to determine strategies.

Investment Management Tech Trends

  • Fund managers will soon be required by law to provide much more detailed analysis of individual investments, requiring a move from siloed to holistic systems that can capture the full breadth.
  • Asset managers need more analytics tools that help find alpha, correlations, and create visualizations on large data sets to meet increasing big data demands and to clarify information.
  • Mid-sized asset managers need to satisfy client demand for transparency and reliability in returns.
  • Automated processes and simplified workflows are necessary to ease reporting requirements and reduce the risk associated with the due diligence process.




What does Big Data mean to these firms?

  • A way to get a complete and transparent view into their investment world.
  • Usability à Advisers need to be able to quickly harness, analyze and report on big data

“Due to increasing market and regulatory complexity with mandates such as EMIR, Dodd-Frank and FATCA, asset managers have to consistently demonstrate transparency and better disclosure to mitigate risk. A key element of effective risk management is to have a holistic view of information before investment decisions are executed. The ability to focus on core competencies and new delivery models that increase efficiency and lower total cost of ownership will differentiate fund managers from their competitors and help grow assets under management.”

Quote from a Director at a top Financial Research Analyst Firm

What makes Authentic Intelligence different?


  • We offer a dynamic and unlimited learning model that adapts, evolves, and explores – far more than simply a technical trading system can offer.


  • We utilize a continuous-learning framework that maximizes asset allocation effectiveness and is only limited by the processing power of its operational environment


  • Our unique end-to-end life-cycle collects data, analyzes patterns, strategizes, simulates, implements asset reallocations, and discovers new data sources.


  • We offer an automated patent-pending web-discovery component triggered by a continuous feedback loop from strategy/simulation agents to spawn dynamic web data search, discovery, and cataloging. These values and parameters can then be seamlessly integrated into the investment models and retested in a continuous improvement fashion that constantly discovers, catalogues, and evaluates new data sources for inclusion into the model.


The Authentic Intelligence Process

A platform used for analyzing the equities market, discovering market patterns, executing simulations to test and refine strategies, and deploying strategies into the live market.


  • Leverages multiple redundant data feed services to integrate quote information, equity fundamentals, and technical indicators into a unified database.


  • Automates the collection of this vast information along with providing a dynamic method for scraping the web for market information not available through standard data feeds.


  • Automated data movement features that allows for unlimited data access including the integration of both technical and fundamental data


  • Assesses the value and reliability of internal and external research, and dynamically discovers and validates new meaningful data sources to improve and streamline information flow.

The Authentic Intelligence Process (continued)


  • Information is stored and analyzed for optimal trading patterns through both analytical queries/reports as well as through trading simulations to test various strategy models.


  • Includes a patent pending technique to enable automatic learning not only from analyzing data, but from analyzing simulation processing to learn how to recognize leading indicators in the market and continually improve strategies and parameters associated with a customer’s investment strategy.


  • To help account for unexpected market events and changes in macro-economic dynamics, Authentic Intelligence dynamically sifts through massive amounts of market data to find the most pertinent information and then is able to create new validated investments models and/or optimize and improve existing investment models automatically based on new real-time market data and indicators.


Scenario 1 – Web Scraping Process


  • Identify a web site/data source that has desired information – i.e. Consumer Credit Outstanding by month –(Federal Reserve Releases)


  • Define starting tag to identify table on the page and the columns – i.e. – Define the following tabs in the web data download metadata tables –


  • Identify any query string required to drive the data – for example: if downloading insider sales, this would be driven by equity symbols. For global-type data such as consumer credit, there would be a static index defined such as “Fed” with the metric defined as CCO – “Consumer Credit Outstanding”


  • Generate technical indicators against the CCO including moving average, relative strength – utilize smoothing to granulize down to a daily level metric so that the indicators can be computed on a daily schedule.


  • Once the data has been enabled for the daily web download process, the metrics will be populated and available for correlation analysis on the Power Views.


  • Demonstrate the values and how they correlate to monthly performance of an index such as the S&P using the Power View visualizations.



Scenario 2 – Incorporating metrics into a trading strategy


  • Utilize scenario 1 web scraping process for NYSE Margin Debt – http://www.nyxdata.com/nysedata/asp/factbook/viewer_edition.asp?mode=table&key=278&category=8


  • Incorporate existing metrics for Bollinger band, RSI, and volume to integrate parameters that indicate overbought/oversold conditions as triggers


  • Utilize scenario 1 approach to generate technical indicators including Bollinger band values


  • Integrate margin debt metrics such with a parameterized strategy that tests for extremes where the standard deviations are violated on the bands in order to create short or long positions against the S&P 500 with parameterized entry/exit values.


  • Use intraday SPY to check for high or low of day reached after 12 PM as a trigger to enter a position.



Scenario 3 – Reports & Analytics


  • Utilizing trading strategy for scenario 1, display profitability reports including reports for auto-learning strategies (super-strategies that evaluate the effectiveness periods for employing the underlying strategy).


Scenarios 4 & 5 – Trading Interface


  • Setup paper trading account with Interactive Brokers


  • Fire up live-trader interface


  • Ensure daily simulator is running


  • Setup the trading strategy from scenario 2 to run after 12 PM and utilize intraday near-real-time scraping to check for S&P hitting low or high of day and integrate with daily strategy parameters for margin debt, Bollinger band, volume, RSI.


  • Generate a report for daily recommendations based on the strategy and output to a spreadsheet.


  • Import the spreadsheet into Collective 2 in order to perform trades.



Scenario 6 – Predictive Analytics

  • Utilize Predictive Analytics to run against the strategy results data in order to determine correlations and provide confidence intervals



Scenario 7 – Advanced Reporting & Predictive Analytics


  • Utilize Predictive Analytics to run against the strategy results data in order to determine correlations and provide confidence intervals


  • Demonstrate Power Views that show the impacts of various indicator/indices and how they correlate to next day behavior


  • Create new Power View that follows approach of Power View #1, but does this against actual profit results based on back-testing.



Multi-Server-based architecture

Utilizes parallel agents to run simulations that validate virtually any investment model against comprehensive database of auto-adjusted market valuation history.

The optimizer can scale to thousands of concurrent simulations through a massively efficient server infrastructure based on PCIE SSD.

Integration to statistical predictive packages

Calculates the reliability coefficients for particular models.

Big data is intelligently shrunk to more meaningful smaller datasets through a patent-pending pattern recognition process that pre-stages analytics processing.

Automated patent-pending web-discovery component

Triggered by continuous feedback loop from strategy/simulation agents to spawn dynamic web data search, discovery, and cataloging.

Collected data becomes part of the investment model database.

These values and parameters can then be seamlessly integrated into the investment models and retested in a continuous improvement fashion that constantly discovers, catalogues, and evaluates new data sources for inclusion into the model.

Patent-pending auto-learning technique

Continually evaluates performance of the entire system.

Automatically implements improvements for the techniques used to analyze investment model effectiveness and calculate viabilities of models.



Database Attributes

Financial Market Trend Analysis

1 billion+ intraday rows loaded daily and analyzed

Generation of numerous key statistical indicators

166 million quotes

300 million quote metrics

Rolling correlation patterns to identify predictive relationships

Thousands of simulations to test impact of values changes on several different indicators


Identify significance of indicators and how significance changes over time

Run simulation in an on-going basis in order to interact directly with financial market to mitigate risks and maximize profits.

Ongoing analysis of decision-making monitoring variables and effects of dynamic adjustment to evolve strategies.

Direct market monitoring and implementation of recommended actions from Simulator “Autolearn”


Microsoft Opportunity

Authentic Intelligence leverages the Microsoft platform extensively.

The server aspect of the system runs on Windows Server 2012 64-bit software using SQL Server 2012.

All application code is developed in Visual Studio.NET and all development work is controlled through a Visual Studio Team Foundation (TFS) server.

SQL Server Integration Services (SSIS) is leveraged for much of the data loading processes.

The system includes a Windows service developed in .NET to support continuous simulation and analytical processing.

A service-oriented architecture through the use of Windows Communication Foundation to supports web interfaces from lightweight clients.

The client aspect also leverages the Microsoft platform using Windows .NET forms for the Automated Trading component that interfaces through a Java wrapper class written in .NET to perform automated trading with the Interactive Brokers online brokerage system.


Technical Capabilities

Intelligent Data Processing

Turns large amounts of data into smaller more meaningful data.

Simulation including autonomous learning (“Auto-learning”)

Implements continuous feedback to improve not only the strategies but the strategy selection process itself.

Live Trading

Implements the strategies tested and analyzed to be best into real trades using the Interactive Brokers API.

Performance reports

Effects of parameter variations for the strategy and the impact of different auto-learning intervals.


How any set of indices and metrics affect the future performance of any other set of indices/equities

External Data Integration

Dynamic web-spidering technology to capture non-obvious metrics to analyze and incorporate into strategies

Intelligent Data Processing

Loads over 4 million intraday rows per day in just a few minutes and roll-up the time-of-day significant events.

Uses the time of the day for high and low and volumes which helps to ensure the simulator is realistic for evaluating the fill price.

Automatically adjusts splits and dividends on a daily basis to improve simulation quality.

Uses multiple data sources as integrity checks – both data from Yahoo finance as well as paid subscription service to validate price history.

Pattern recognition – Pre-aggregates data as it arrives leveraging PCIE-SSD to identify behaviors across multiple equities and indexes.

Reduces dataset sizes dramatically to allow more advanced correlation analysis.

Intelligent Data Processing

Over 1.2 billion intraday rows and several tables with dozens of millions of rows

Utilizes PCIE SSD to achieve over 5 GBPs throughput

Split/dividend adjusted back to stock origination date








Pattern Recognition techniques

Significant Events

Utilizes patent-pending technique to compress market history into simple integers and then identify correlation including offset-correlations instantly.

Isolates the trend-reversal points and not the actual prices






Aggregates the trend-reversal points into a time-series bit mask

Ranks based on the ideal patterns rather than all transactions.




Uses high-quality end-of-day historical data for equities with split and dividend adjustments back to the issue origination date.

Includes optional data back to January 2003 as well as all major indexes and currency pairs.

Supports 15-minute interval simulations using Intraday data going back to January 2009 with potential to support 1-minute intervals.

Allows any index, equity or indicator to serve as a trading trigger including non-market metrics gathered from external sources such as Fed actions, CPI, and unemployment data.

Integrates auto-learning to continuously evaluate portfolio performance linked to parameters and dynamically change parameters linked to a live trading portfolio.



Simulation Walk-through

Select parameters to utilize from metrics and test ranges








System generates portfolios



Simulation – Interval Setup

Allows dynamic definition of parameters based on analytics processing to identify most likely values to yield positive ROI


Simulation Walk-through

Generate portfolios with static parameters, then auto-tune using auto-learning


Simulation with Auto-learn Impact

Live Trading

  • Dynamically selects the most successful portfolio from the simulation portfolios using criteria that is constantly refined based on Auto-learn.
  • Generates higher-level simulations applying different profitability windows for success criteria.
  • Integrates metrics dynamically into the simulations selected via analytic processing which features external data collection.
  • Interfaces with Interactive Brokers API for live or paper-trading accounts supporting order techniques including One-Cancels-Another (OCA) for combined stop/limit orders, market timing parameters, and other standard trade parameters.

Performance Reporting

  • Analyzes strategies and the effects of parameters and their values.
  • Tracks performance of both simulation and live portfolios with results updated in real time.
  • Able to quickly estimate the ROI of a strategy through the “what-if” analytic mode without the need to run a complete simulation.
  • Identify risk versus reward and volatility of different strategies including the effect of auto-learn windows.


Show impact of any combination of indicators with metrics on another set of indicators and their metrics. Indicators may include non-trading instruments such as CPI, unemployment, or any custom indicator provided through the external data capture feature.

Metrics linked to indicators include technical items such as RSI, moving averages, as well as fundamental items or items specific to the indicator.

Rank the probabilities that one set of indicator/metrics correlate to future performance over different time intervals on a trading instrument.

Utilize predictive analytic tools to identify optimal triggering indicator/metric combinations along with confidence intervals (user process converting to be done by system) .

Incorporate discoveries from metrics back into strategies – (currently manual process, in process of automating).


Data Discovery

Provide extensibility of the database to incorporate any indicator or metric available in a data source including those hosted by government web sites such as the Fed or Bureau of Labor Statistics

Indicators or metrics are analyzed and then persisted into a generic metrics table that all strategies are able to access and exposed to the visualization tools

Examples of indicators/metrics beyond traditional market indicators include CPI, insider trade data with plans to include Fed actions, unemployment statistics, and social media trends


Visualization Examples