SAFIN(FRIE)++ : type-I online mandami fuzzy inference system with application in option trading
Fuzzy neural networks are often used to handle dynamic data stream in the financial market. However, unlike stock data that is updated every tick, option data is often sparse in nature. To handle the sparsity in data set, a fuzzy neural network system requires interpolation and extrapolation feature...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project |
| Language: | English |
| Published: |
2017
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/72908 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | Fuzzy neural networks are often used to handle dynamic data stream in the financial market. However, unlike stock data that is updated every tick, option data is often sparse in nature. To handle the sparsity in data set, a fuzzy neural network system requires interpolation and extrapolation feature.
This paper employs interpolation and extrapolation techniques of [1] to SAFIN++ system in paper of [3] to improve the accuracy of the system when concept drift and shift are detected or when the rule base of the system is sparse.
This paper proposes a novel neural fuzzy system architecture called SAFIN++ with Fuzzy Rule Interpolation/Extrapolation that has the following features:
a) Online learning feature
b) Capable of detect concept shift or drift and handle concept shift or drift using interpolation or extrapolation using multiple multiple-antecedents fuzzy rules
c) Rule forgetting or decay memory |
|---|