| The main objective of the EUROSIGNAL project is the development of a software prototype incorporating a large-scale real-time simulation model of the stock market that will be capable of providing useful predictions of future short and medium term trends in stock exchanges, in several groups of stocks and in individual stocks. The targeted group of users for such predictions include private and institutional investors, investment banks, investment consultants, stock exchanges and policy makers. The simulation model will include a balanced consideration of macroeconomic, valuation and stock market variables and will provide predictions of probable turning points which is a major challenge for researchers in the economics and finance areas, in view of the complexities involved in today's Digital and Global Economy.
OBJECTIVES
The main objective of the EUROSIGNAL project is the development of a software prototype incorporating a large-scale real-time simulation model of the stock market that will be capable of providing useful predictions of future short and medium term trends in stock exchanges, in several groups of stocks and in individual stocks. The targeted group of users for such predictions include private and institutional investors, investment banks, investment consultants, stock exchanges and policy makers. The simulation model will include a balanced consideration of macroeconomic, valuation and stock market variables and will provide predictions of probable turning points which is a major challenge for researchers in the economics and finance areas, in view of the complexities involved in today's Digital and Global Economy.
DESCRIPTION OF WORK
As already mentioned, the main objective of the EUROSIGNAL project is to provide leading indicators to private and institutional investors, investment consultants, investment banks, policy makers and the SEs themselves concerning future trends in specific stock markets. These predictions of future trends will be based on a large-scale real-time simulation of the stock market. A methodology involving a 3-layer approach will be employed, which makes effective and integrated use of the knowledge extracted from a large amount of information originating from recent surveys, historical data, unstructured data from various sources, and current real-time stock transactions.
The first phase of the project involves the further development and use of new data mining techniques, based on tree-classifiers and clustering in order to extract relationships between macroeconomic variables, and to determine which variables have the greatest effect on asset prices.
The second phase of the project involves the pricing of individual assets using two methodologies:
(i) the full anticipated earnings valuation technique developed by ECONET, and
(ii) the Real Options approach.
Both of these methods are based on the idea that a company's valuation is greater than its Net Present Value, and should also include the value of future possibilities of exploiting favourable market conditions. Approach (i) requires inputs from the first phase, whereas approach (ii) requires inputs from both the first and third phase below.
The third phase of the project involves:
(i) The modelling of asset price dynamics using both linear and non-linear stochastic models and adaptive prediction with blind equalisation techniques,
(ii) The further development of advanced simulation techniques based on new low-discrepancy sequence sampling,
(iii) The use of the results in (i) and (ii) above to generate arbitrage-free state space transition graphs of the economy, particularly of the stocks listed on a SE, thus generating predictions for the SE index, sector indices, and for individual stock prices.
EUROSIGNAL is designed to become a major player in stock market predictions and it is envisaged that customers will include: Private and institutional investors, Investment banks, SEs, Investment consultants and Policy makers. The project involves RTD for the construction of a software prototype incorporating a large-scale simulation model of an SE, which will be able to provide in real-time useful predictions of short and medium term trends in the SE Index and in individual stock prices. |