SVTbx: Security Valuation Toolbox
Last update: June 2024
"Beware of geeks bearing formulas." [Warren Buffet]
In 2015, I developed a Security Valuation Toolbox. People value businesses, computer software does not. Using computers can and does help with reporting, automatizing routine procedures, and some analysis, though. They won't read a 10-K report for you but can do other useful stuff.
In 2025 the Toolbox is getting a major re-haul -- it is being ported to Python ecosystem, uses larger and more refined dataset and leverages ideas from economics, finance, information theory, and machine learning. Currently the toolbox can sift through thousands of companies on more than 20 world exchanges. Pictures below are from the "old" system...
The toolbox uses detail accounting and other market data to analyze publicly traded companies and helps to challenge the market price. Advanced screening and "intelligent" watch lists are an important part of the toolbox. The purpose of the SVTbx is informational only and does not offer a substitute for a thorough analysis of a business, its financials, and competitive position.
The valuation techniques used are based on my own analysis, relying mostly on accounting valuation methods, asset pricing, and information theory, with the help of computational statistics and machine learning. The toolbox goes beyond simplistic screening of basic ratios and quickly applies a thorough analysis on each company, markets, and portfolio composition.
This is mainly a research project testing new approaches to top-down and bottom-up monitoring and surveillance of financial markets, potentially useful for "early warning exercises."
COVERAGE:
The core of the SVT analysis uses data for businesses traded on selected markets, namely the U.S. market (NYSE, NASDAQ, AMEX and Pink Sheets for some applications), Frankfurt/XETRA Stock Exchange, Paris, Brussels Euronext, Australia's ASX, Toronto Stock Exchange, London's LSE, Stockholm, Oslo or India stock exchanges. The space covers roughly 10K+ companies, excluding Pink Sheets. I use data for key financial and accounting indicators (balance sheet, income statement, and cash-flow statement data) and information about prices, volume traded, ownership, industry, etc.
PLATFORM:
The toolbox is implemented using Python-MySQL-LaTeX-JavaScript/HTML environment with some old Matlab/Java components for the most part, using a codebase that is easy to port to other systems. Reporting is based on HTML/JavaScript and pdfLaTeX/TikZ (PDFs). The software is for personal use and all the examples are only illustrative. Please, note that none of these reports constitute an investment advice.
Examples of Use:
Notes & Research Papers
"Beware of geeks bearing formulas." [Warren Buffet]
In 2015, I developed a Security Valuation Toolbox. People value businesses, computer software does not. Using computers can and does help with reporting, automatizing routine procedures, and some analysis, though. They won't read a 10-K report for you but can do other useful stuff.
In 2025 the Toolbox is getting a major re-haul -- it is being ported to Python ecosystem, uses larger and more refined dataset and leverages ideas from economics, finance, information theory, and machine learning. Currently the toolbox can sift through thousands of companies on more than 20 world exchanges. Pictures below are from the "old" system...
The toolbox uses detail accounting and other market data to analyze publicly traded companies and helps to challenge the market price. Advanced screening and "intelligent" watch lists are an important part of the toolbox. The purpose of the SVTbx is informational only and does not offer a substitute for a thorough analysis of a business, its financials, and competitive position.
The valuation techniques used are based on my own analysis, relying mostly on accounting valuation methods, asset pricing, and information theory, with the help of computational statistics and machine learning. The toolbox goes beyond simplistic screening of basic ratios and quickly applies a thorough analysis on each company, markets, and portfolio composition.
This is mainly a research project testing new approaches to top-down and bottom-up monitoring and surveillance of financial markets, potentially useful for "early warning exercises."
COVERAGE:
The core of the SVT analysis uses data for businesses traded on selected markets, namely the U.S. market (NYSE, NASDAQ, AMEX and Pink Sheets for some applications), Frankfurt/XETRA Stock Exchange, Paris, Brussels Euronext, Australia's ASX, Toronto Stock Exchange, London's LSE, Stockholm, Oslo or India stock exchanges. The space covers roughly 10K+ companies, excluding Pink Sheets. I use data for key financial and accounting indicators (balance sheet, income statement, and cash-flow statement data) and information about prices, volume traded, ownership, industry, etc.
PLATFORM:
The toolbox is implemented using Python-MySQL-LaTeX-JavaScript/HTML environment with some old Matlab/Java components for the most part, using a codebase that is easy to port to other systems. Reporting is based on HTML/JavaScript and pdfLaTeX/TikZ (PDFs). The software is for personal use and all the examples are only illustrative. Please, note that none of these reports constitute an investment advice.
Examples of Use:
- Elementary valuation screen example (printout w/o interactive features; see figure below for an older version)
- "Moat Watch" Screen Example [pdf] -- a printout of a JavaScript-powered report w/o interactive features
- Examples of recursive valuation (ALL) using three valuation scenarios...
- Basic analytical report for a company and 'intrinsic value' computation
- Relative position screening (screen based on reporting firms which for N parameters are better than median of its peers)
- ...
Notes & Research Papers
- R1 Valuation Model -- structure, assumptions, uncertainty assessment
- How to Gamble If You Must and If You're Not Gambler
Example of a valuation screen. D/E = Long-Run Debt to Equity, Dist2Low = P/52weekLow, Value = result from the SVTbx valuation, with the value in the "pop up" bubble (V_SR) being a more conservative valuation for the particular company.
Tracing Market Dynamics: Histogram of Price-to-Value Ratio of Pre-Screened Companies
**) Histogram of approximate price/value ratio for companies with long-term debt share of equity smaller than 0.45, history of positive free cash flow, and book value CAGR (after adding back dividends) at least 7%. N = 504. Markets: Nyse, Nasdaq, Deutsche Borse/Xetra, ASX, TSX, Euronext Paris, Stockholm, Oslo.
Pseudo Real-Time Valuation Benchmarking Example: Woolworth Ltd., Australia
The three valuations are relevant for both their level and their dynamics and aim to challenge the market price. The dynamics provides the information about the suitability of the model setup for the corporation considered. This is a way to find out if the valuation is too optimistic on average, for instance. The level of three valuations depends on different assumptions and thus provides some information of valuation uncertainty. "X" valuation is rather extrapolative of about current situation, "M" value is more conservative and makes assumptions about different stages of firm and industry maturity, while "RHO" valuation is a rather defensive valuation, an envelope, of a situation where the firm loses quickly its competitive advantage...
DISCLAIMER:
The views expressed herein are those of the author and should not be attributed to the International Monetary Fund, its Executive Board, or its management. The toolbox is for private use and research purposes only, based on publicly available information.
The views expressed herein are those of the author and should not be attributed to the International Monetary Fund, its Executive Board, or its management. The toolbox is for private use and research purposes only, based on publicly available information.
