PresentationPDF Available

PhD Thesis Presentation: MODERN METHODOLOGIES FOR BUSINESS INTELLIGENCE SYSTEMS DESIGN

Authors:
  • Algorithm Invest

Abstract

PhD Thesis Presentation Moredn Methodologies for Business Intelligence Systems Design
MODERN METHODOLOGIES FOR
BUSINESS INTELLIGENCE SYSTEMS DESIGN
Ph.D. thesis
Author: Cristian Mihail C. Păuna
Scientific conductor: Prof. Univ. Dr. Ion Lungu
Economic Informatics Doctoral School
Economic Studies Academy, Bucharest, Romania
Knowledge providers:
Economic Studies Academy, Bucharest, Romania
World Academy of Science, Engineering, and Technology
Technische Universität Graz, Austria
Alexandru Ioan Cuza University, Iași, Romania
West University, Timișoara, Romania
Algorithm Invest (algoinvest.biz)
Business Intelligence Systems
are the sum of strategies, information
technologies and systems to analyze the
past data in order to optimize the
business decisions of today to generate
better results in the future
In today's real-time Business
Intelligence including expert and
automated decision-making systems
we have 4 conceptual levels
=> human decision =>
past
data
better
future
=> automated decision =>
past
data
optimal
results
real
time
data
+
The informational system represents the totality of the information flows, methods,
tools, and techniques to collect, store, and process data for decision elaboration,
transmission, and transformation into results.
The informatics system is a structured set of interconnected functional elements
necessary to ensure the process automation of obtaining information, substantiating,
transmitting, and transforming decisions into results.
Real-time
Business Intelligence
System
Trading models,
strategies and systems
Economic
Informatics
Cybernetics
Time delay
as design factor
for real-time
Business intelligence
Systems
Time delay
as design factor
for real-time
automated capital
investment systems
Trading
system
types
Investor's
trading
system
"By 2017, 60% of the big data projects
will NOT be able to pass the experimental phase and will be abandoned" (Gartner, 2015)
ProposedProgressMproject management methodology for real-time Business Intelligence Systems
ProgressM
component
Iteration to
optimize a
key factor
C1 C2 C3 ...
Automated capital investment informatics system
Parallel calculation procedures
in automated capital investment informatics system
Time-series indicators cube
Business intelligence comes from:
- real-time data analysis
- low-latency data processing
- profitable data-mining methods
- parallel computational models
- real-time machine-learning
- time-series indicators cube
Only open source technologies used
in my automated capital investment system
FreeBSD 12.1 UNIX servers since 1993
Apache 2.4 http servers since 1995
PHP 7.4 programming language since 1995
JavaScript programming language since 1995
MySQL 8.0 relational database since 1995
C++ programming language since 1990
Python programming language since 1990
node.js runtime environment since 2010
MQL4 programming language since 2008
MetaTrader4 trading platform since 2008
Contributions in real-time
Business Intelligence Systems design
methods and methodologies
ProgressMproject management
methodology for real-time BIS
SignalM method to automate investment signals
OptimM method to optimize investment strategies
ModelM method to automate an investment
model
Improvements in well-known
capital trading and investment
strategies and models
using Fisher Transform
using Inverse Fisher Transform
using Smoothed Heikin-Ashi Transform
using Trigonometric Interpolation
using original Turtle System
using Oversold Price Levels
using multi-levels low-risk entry levels
Contributions as new methods in
capital trading and investment
automated software systems
Global Stop-Loss Method to automate
risk and capital exposure management
Logarithmic Risk Distribution to build
a stable long-time capitalization
Limit Conditions for capital markets
to optimize entries and exits events
Price Cyclicality Function
Volume Cyclicality Function
Price Probability Predictor
Price Prediction Line
Price Prediction Bands
Inverse Fisher RSI
Silent Market Indicator
Quality Trading Coefficient
Wait-On-Signal Method
Best paper award, April 2019, Athens, Greece
Best paper award, September 2019, Rome, Italy
Management & Marketing Award
March 2019, Bucharest, Romania
"Low-risk trading algorithm based on the
price cyclicality function for capital markets"
Between 2017 and 2020:
24 research papers
14 scientific articles published
10 international conferences attended
Limit conditions types in financial markets
1. Don't buy near
maximum price levels
2. Don't trade on
non-significant price movements
3. Don't invest in
low price probability
Price Prediction Line
Using Price Cyclicality Function (PCY)
Using Inverse Fisher RSI Function (IFR)
Using Price Prediction Line (PPL)
Using Silent Market Indicator (SMI)
Using Price Probability Predictor (PPP)
Silent Market Indicator
Price Probability Predictor
Time delay results
Computational delays
from 120 to 380 msec
Data connection delays:
from 50 and 600 msec
Total delays obtained:
from 170 to 980 msec
Capital investment results in 659 days
between 01.07.2018 and 20.04.2020,
on 17 international markets:
10288 entry signals, 5527 trades,
0.18% daily profit, 0.90% weekly,
5.42% monthly profit, 65% yearly
Thousand thanks to:
Prof. Univ. Dr. Ion Lungu
Prof. Univ. Dr. Mihai Roman
Prof. Univ. Dr. Paul Pocatilu
Prof. Univ. Dr. Adela Bâra
Prof. Univ. Dr. Iuliana Botha
Prof. Univ. Dr. Vlad Diaconița
Prof. Univ. Dr. Marian Dârdală
Prof. Univ. Dr. Mihaela Muntean
Prof. Univ. Dr. Dinu Airinei
Prof. Univ. Em. Dr. Ing. Corneliu Berbente
Prof. Florin Vulpescu-Jalea (1952-1991)
Prof. Univ. Dr. Ion Gheorghe Roșca (1949-2013)
Modern Methodologies for
Business Intelligence Systems Design
Ph.D. thesis
Author: Cristian Mihail C. Păuna
Scientific conductor:
Prof. Univ. Dr. Ion Lungu
Email: cristian.pauna@ie.ase.ro
Phone: +407.4003.0000
Economic Informatics Doctoral School
Economic Studies Academy
Bucharest, Romania
ResearchGate has not been able to resolve any citations for this publication.
ResearchGate has not been able to resolve any references for this publication.