27-10-2016, 03:06 PM
CERTAIN INVESTIGATIONS ON FUZZY METAGRAPH BASED DATA STRUCTURES AND ITS APPLICATION TO DECISION SUPPORT SYSTEM FOR ANALYSIS OF INDIAN STOCK MARKET
1461753608-vivaaprthiru20162.ppt (Size: 2.34 MB / Downloads: 6)
ABSTRACT
This research deals with the Fuzzy metagraph (FM) based data structures to help decision making and reducing risk in share market investment for short term and medium term duration.
Investing in stock market is not an easy task. It needs careful decision making skills to maintain profit in the long run.
Relative Strength Index (RSI), Moving Average Convergence Divergence (MACD), William- %R, Stochastic Oscillator (SO) and Negative Volume Index (NVI) are some of the Technical Indicators which are used as input to train the system which is integrated with Fuzzy Metagraph.
The Decision Support System (DSS) is optimized using Genetic Algorithm (GA), the GA optimization improves overall profit by 30% to 40%.
Stocks listed in Bombay Stock Exchange (BSE) in India are used to evaluate the performance of the system.
1. Introduction
Graphs play an essential role in the design of most of the information processing system. Data about Data is metadata like graph about graph may be metagraph.
A graph is Abstract Data Type (ADT) which is meant to implement the metagraph concepts from graph and hyper graph.
Some aspects of a graph theoretic problem may be uncertain. It is natural to deal with the uncertainty using the methods of fuzzy sets and fuzzy logic.
Fuzzy Metagraph (FM) is a graphical hierarchical data structure for defining directional relationships between sets of (one or more) elements. It has all the properties of fuzzy graph and metagraph.
FM is an emerging technique used in the design of many information processing systems like transaction processing systems, decision support systems, and workflow Systems.
1. Introduction – Cont.,
An equity market is a public entity for the trading of company shares. The purpose of a stock exchange is to facilitate the exchange of shares between buyers and sellers.
A trader decides about what share to trade, when to trade, using either fundamental or technical analysis.
Generally in stock markets, most of the investors lose their capital money because of dynamism and unpredictable environment of stock market domain.
Investing in stock market is not an easy task. It needs careful decision making skills to maintain profit in the long run.
The Timing Problem to buy low and then sell high is a non-trivial problem and it is considered a dream of each investor.
1.1. Problem Definition
Some of the problems involved in investing are;
Taking decision which stock will be appropriate to buy or sell for better profit.
Analyze and extract useful information from various technical indicators available
in order to make qualitative investment decision.
Having a reliable Decision Support System (DSS) which is fast and reasonable accurate
To investigate graph based data structures to provide solution for decision making in stock market.
Hence, FM-based DSS model is proposed to overcome the problems stated above.
1.2.Objective
The goal of this research work is to develop a DSS model for decision making in stock market with minimal risk.
Study the usefulness of FM based data structures in designing an effective DSS
Develop a FM based DSS model to arrive an accurate decision making in stock market investments.
Optimize the DSS model to improve its performance
Compare its performance with available existing methods.
1.3.Graph Data Structures
Minimum Spanning Tree (MST) is a graph optimization method, which has been constructed from general graph. A single node can’t store more than one data in MST, so to go for higher level graph data structure.
A hypergraph is a graph in which more than two vertices are linked by the same edge. A partial solution is offered by hypergraph, it means the expansion of graph models for the modeling complex system.
Hypergraph Based Data Structures (HBDS) are built on four fundamental concepts, which taken together are sufficient to describe any real world phenomenon like object, class, attribute and relation.
According to nonlinear tree data structures, like B tree and B+ tree consist of one or more data in a node but it is not applicable in other tree structures and also applicable in metagraph structures.
1.4. Metagraph construction from general graph
If a graph contains cycle or loop then merge the loop of vertices into single node or metagraph vertex. The graph is transferred to metagraph with the help of clustering method.
Data about Data is metadata like graph about graph may be metagraph.