News

citefactor-journal-indexing

Fuzzy Aggregate Candlestick and Trend based Model for Stock Market Trading

Forecasting stock markets is always fascinating. In this paper we have used fuzzy rule base and approximate reasoning for creating a model that can be used to predict stock market trend and forecast crisp values. A unique approach is implemented by creating a single aggregate candle from a range of observations and that aggregate candle is fuzzified such that it represents the sentiment of the traders. The concept of time windows is used to generate fuzzy antecedent and consequents. Using Mamdani implication and fuzzy inference mechanism the future trend is forecasted in fuzzy terms and after which crisp futuristic value is generated as the next target for the stock market. The final output from our proposed system is in the form of decision rules that would suggest which Action (buy/sell) the trader should take and for how much Target (crisp value) the trader can expect after taking the specified action in the next trading session. The forecasting efficiency of the proposed model is calculated in the terms of RMSE and compared with other benchmark models the proposed fuzzy model displays promising results. Transaction data of CNX NIFTY-50 index of National Stock Exchange of India is used to experiment and validate the proposed model



Real Time Impact Factor: Pending

Author Name:

URL: View PDF

Keywords: Approximate reasoning, Candlestick chart, Fuzzy logic, Prediction, Time series

ISSN:

EISSN: 2320 – 6527

Add Citation Views: 1














Search


Advance Search

Get Eoi for your journal/conference/thesis paper.

Note: Get EOI for Journal/Conference/ Thesis paper.
(contact: eoi@citefactor.org).

citefactor-paper-indexing

Share With Us












Directory Indexing of International Research Journals