The EURO BROAD INDEX tracks the returns of a portfolio of stocks chosen daily from more than 2500 equities traded in the Paris, London, Frankfurt, Milan and Brussels stock exchanges. This dynamic rebalancing portfolio is determined by valuing 40 company indicators on 5 ensembles, using a data driven fuzzy logic scoring method. Our stock dashboard offers analysts a framework to increase their market awareness and allows investors to develop value based long/short equity investment strategies (check out our initial list).

  • Overview
  • Method
  • News
  • FAQ
  • Products & Services
  • R & D
  • Partners
  • Contact

CARTE BLANCHE EURO BROAD INDEX is a dynamic rebalancing virtual portfolio system, tracking the daily performance of equities traded in the stock exchanges of Paris, London, Frankfurt, Milan and Brussels. The index is determined by comparing company financial ratios on five distinct asset ensembles:



We consider nine broad super Sectors:



And define ensembles by market capitalization as:

// LARGE CAP > €/£ 5 billion    // €/£ 1 billion < MEDIUM CAP < € 5 billion    // SMALL CAP < €/£ 1 billion

The index describes the evolution of the portfolio weighted by score logarithmic returns calculated at closing prices. Stocks with a recommendation of outperforming or buy are picked and weighted according to our Fuzzy Logic scoring method. Modern applications have shown that control of complex systems is improved when dynamics are controlled by a Fuzzy Logic set of rules. The reasoning is straightforward. Fuzzy Logic Systems assume the existence of uncertainty and are able to adapt and evolve according to the dynamics of the system it controls. Tracking logarithmic returns, on the other hand, provides a reasonable proxy to account for trading costs compared to real returns. Gains are slightly discounted and losses are slightly amplified thus offsetting the transactions costs associated with active portfolio management.

The virtual trader assumes that transactions are performed the next trading day at the previous closing prices. For example, returns from a closed position are accounted for in the next trading day and returns from a new position start being accounted for in the next day. The index can thus be thought of as a historic account of our portfolio choices, whereas our stock lists, depicted in the stock dashboard are strategic decisions to be undertaken in the future (check out our initial list).

Our data covers 40 balance sheet, income statement and cash flow ratios. Each company daily indicator evolution is scored in fuzzy sets defined by statistical measures of historic market data for each ensemble. Our objective is to accommodate in our sets both cyclical and outlier market phenomena. Assets in each ensemble are then given a score from 0 to 1000. To obtain our BROAD SCORE, each indicator is weighted equally within each ensemble, while ensembles are weighted as:

// SE (10%)    // SEC (2.5%)    // SAE (2.5%)    // SAC (2.5%)    // SAS (82.5%)


// Price to Earnings (ps,ttm)     // Price to Book (ps,mrq)     // Price to Revenue (ps,ttm)     // Dividend Yield (ps,ttm)

// Operating Margin (ttm)     // Operating Profit (ttm)     // Return on equity (ttm)     // Return on Assets (ttm)

// Price to Free Cash Flow (ps,ttm)     // Enterprise Value(EV) to EBITDA (ttm)     // Payout Ratio (ttm)     // EV to Gross Profit (ttm)

// Earnings to Revenue (ttm)     // Earnings to Gross Profit (ttm)     // Quarter Revenue Growth (YoY)     // Quarter Earnings Growth (YoY)

// Debt to Equity (mrq)     // Operational Cash Flow To Debt (ttm,mrq)     // Cash Available For Debt Service (mrq)     // Current Ratio (mrq)

// EV to Revenue (ttm)     // Price to Cash Flow (ps,ttm)     // Price to Available Cash (ps,mrq)     // Price to Gross Profit (ps,ttm)

// Operating Ratio (ttm)     // Price to Net Income (ps,ttm)     // Free Cash Flow to Revenue     // EBITDA Debt Coverage Ratio (ttm,mrq)

// Net Debt to Book Value (mrq)     // Gross Margin     // Operational Cash Flow to EV yield (ttm)     // Free Cash Flow to EV yield (ttm)

// Price to Operating Income (ps,ttm)     // Revenue Cost Ratio (ttm)     // Price to Net Cash (ps,mrq)     // Gross Return on Equity (ttm,mrq)

// Net Accrual Equity Ratio     // Price to Non Operating Cash Flow Revenue (ps,ttm)     // Price to Non Operating Free Cash Flow Revenue (ps,ttm)

// Cash flow to Earnings Conversion Ratio (ttm)


// ps: per share     // ttm: trailing twelve months     // mrq: most recent quarter     // YoY: year over year

Our datasets are the result of a continuous integration and audit process of company data obtained from various sources. Our audit processes ensures the quality of the data on a daily basis. Larger data audits are performed periodically to improve our database accuracy and may introduce revisions when these are considered crucial. After auditing, all our data is processed as is. We do not use interpolating methods to complete or modify our databases.

Fuzzy Logic is a tool that can be used to express the inherent uncertainty of the complex systems driving nature. As probability, fuzzy logic can be used to represent subjective beliefs. The main difference between the two approaches relates to the way that uncertainty is modeled. Probability theory is based on subjective probability theorems, whereas Fuzzy Logic deals with subjectivity following the theorems of possibility theory.

Modern probability theory and Fuzzy Logic are both extensions of classical logics intended to deal with issues of uncertainty. Wikipedia describes Fuzzy Logic as “… a form of many-valued logic; it deals with reasoning that is approximate rather than fixed and exact”. When “...compared to traditional binary sets (where variables may take on true or false values) fuzzy logic variables may have a truth value that ranges in degree between 0 and 1.”

To compute our indicator scores, we start be constructing fuzzy sets for a given data set, in one of our ensembles. First, we compute historical averages and standard deviations from our datasets. Then we compute membership functions assuming market normal outcomes are described by a normal distribution, centered in the dataset historical average and with a variance given by a measure of the historical standard deviation.

// Fuzzy Sets Data Classification- This step involves classifying the data according to a set of Fuzzy Logic rules. In the graphics below, we portray this operation with GROSS MARGIN historical data for companies traded in the Frankfurt stock exchange. The first graphic shows the fuzzy sets fitting the historical data.

  • // Definition of Membership Degrees- The fuzzy system classifies the companies according to the classification systems described above by attributing string values ‘H’ (High Performance), ‘N’ (Normal Performance), ‘L’ (Low Performance) for the Fuzzy screener. This classification is defined for the last data January 2, 2015 data on GROSS MARGIN.

    To complete our scoring procedure, we apply a Defuzzification process, to obtain a quantifiable outcome between 0 and 1, for each company GROSS MARGIN indicator. The graphic below portrays the result of the Defuzzification procedure. Mouse-over or tap on the circles to get the details on each quote’s GROSS MARGIN and corresponding score for the Frankfurt exchange.

  • // Defuzzification- The system then produces a quantifiable result in Fuzzy Logic, given Fuzzy sets and corresponding membership degrees. In this specific example we apply a Takagi-Sugeno Fuzzy inference system to obtain final scores. This approach is also defined as an Adaptive Neuro Fuzzy Inference System (ANFIS) since it integrates both neural networks and fuzzy logic principles. It has the potential to capture the benefits of both in a single framework. This inference system corresponds to a set of fuzzy IF–THEN rules that have learning capability to approximate nonlinear functions. ANFIS is considered to be a universal estimator.

    May 02, 2016

    // Introduced new dashboard for the CARTE BLANCHE EURO EQUITY BROAD INDEX stock lists that allows users to track performance and filter stocks by capitalization and momentum.

    February 05, 2016

    // CARTE BLANCHE EURO EQUITY BROAD INDEX stock lists daily updates are now available.

    January 04, 2016

    // CARTE BLANCHE EURO EQUITY BROAD INDEX stock lists weekly updates are now available.

    December 31, 2015

    // CARTE BLANCHE EURO EQUITY BROAD INDEX test period is now concluded and the ensemble weights are now settled forward assuming the following values:

    // SE (10%)    // SEC (2.5%)    // SAE (2.5%)    // SAC (2.5%)    // SAS (82.5%)

    November 2, 2015

    // CARTE BLANCHE EURO EQUITY BROAD INDEX visualization tool is launched after 10 months of successful testing (check out our initial list). The client demonstration cross filter dashboard features all database scores and weights for the initial stock portfolio.

    January 2, 2015

    // CARTE BLANCHE EURO EQUITY BROAD INDEX test version tracking daily over 2500 stocks traded in the London, Paris, Frankfurt, Milan and Brussels exchanges is launched.

    We are currently working on a detailed FAQ section for the CARTE BLANCHE EURO EQUITY BROAD INDEX application.

    If you have any relevant question to improve your understanding of our index, feel free to contact us:

    Contact Form>subject: Frequently Asked Questions (FAQ).

    Our automated virtual portfolio trading machines rely on the integration of three modern computing paradigms:
        - Open Source Code;
        - Parallel Computing on Shared Memory Unix Systems;
        - Cloud Based Client Side Visualization Tools.
    This integration allows our technology to be easily transferred and implemented at an organizational level in a cost effective fashion.
    By using Cloud Based Client Side Visualization, we are able to focus server side operations on complex computations and to provide a modern/powerful equity valuation analysis internet based service.
    These features make our product unique both for individual investors seeking an affordable, easy to use solution that allows to increase their market awareness and financial institutions aiming to increase their organizational technology or provide their costumers with state of the art equity research.

    Carte Blanche currently offer the following solutions:

    // For individual investors

    During 2016 we expect to offer a monthly/annual subscription service for indivual investors interested in acessing our stock listings and scoring analysis. This service will comprise a daily update of our score dashboard and future client based tools, including a guarantee that scores are updated daily before the next trading day starts.

    // For institutional clients

    Carte Blanche offers a range of solutions adapted to the modern regulatory and market challenges faced by financial institutions:
        - Implementation of existing products and consultancy as a service support;
        - Software development of existing systems to meet specific organizational challenges;
        - Comercial partnerships for organizations seeking to provide their costumers with our research service;
        - R&D consultancy services for high end projects.

    // For developers

    Currently, we are working on an open source developer version of our fuzzy logic library to publish online. This version will include the source code for the example described in the methods tab and the data required to reproduce this visualization.

    Most of the time and energy dedicated to our stock indexes is spent engaging in Research and Development activities. Namely, the development of new and better open source code, to be able to improve the quality and reliability of our systems and better adapt these to our costumer needs.

    To achieve this goal, we focus on building integrated systems supported by research and development in four distinct areas of expertise:

    // DATA

    Data management is crucial to our operation. The core of our work is focused on the improvement of data quality, reliability, and the efficiency of database operations.

    At Carte Blanche, data management relies mainly on logical operations and linear algebra manipulation of large datasets. This approach has been made possible by modern computers increasing RAM performance. Most of our data management and manipulation is performed with GNU OCTAVE in UNIX machines. GNU OCTAVE as the unique advantage of allowing efficient data manipulations in an extensive, easy to code statistical and mathematical scientific environment.


    Computing efficiency is a valuable commodity. We believe that modern servers are best suited to perform complex data operations, while client side resources should enable network distribution of computing costs. Distributing computing costs allows us to focus our resources exclusively in R & D, product development and improved client experience.

    To be able to handle both large data sets and complex computations simultaneously in an efficient manner, we build our systems control in GNU C/C++, and rely for thread management on OpenMP, a multi-platform shared-memory parallel programming specification for C/C++. Scalability of our systems is achieved with core parallel computing and efficient distribution of RAM based data operations.


    Fuzzy Logic applications have proven its effectiveness and reliability as a systems control tool for a range of autonomous electronic devices. Given its natural applicability in uncertain decision environments, Fuzzy Logic systems provide a clear tool to test quantitatively, the quality of subjective strategies in multi-criteria decision problems.

    At Carte Blanche, we rely on the GNU OCTAVE fuzzy logic toolkit to compute indicator fuzzy scores. This option allows for an efficient integration of our data management system with one of the most extensive and reliable non-commercial fuzzy logic libraries available.

    In the near future, we expect to include in our services best strategic performance indexes. To achieve this insight, we apply modern evolutionary dynamics methods that seek to maximize the performance of multi-criteria strategies. Our tools integrate aspects of game theory, stochastic methods and adaptive dynamics to optimize outcomes in multi-agent co-evolutionary competitive environments.


    Development of Cloud Visualization Client Side dashboard tools plays a central role in the development of our applications. Modern JavaScript libraries D3.js and Crossfilter.Js allow the development of cloud friendly functional applications.

    These applications have several advantages. Only a limited amount of the clients web browser RAM is used and server-side requests are constrained to a minimum. Integrated D3.Js and Crossfilters.js cloud based dashboards enables the development of fast responding, dynamic easy to use web interfaces. Check out our demo dashboard to see what we are talking about (check out our initial list).

    We are actively seeking to engage in partnerships that help us grow our fuzzy logic market applications successfully.

    If your institution is interested in developing comercial and/or research projects relating to this application, please reach us using our contact form and choosing Partnership & Business as subject.