- We will send in 10–14 business days.
/images/books/13188106/9783639402308.jpg)
- Author: Antonio Stoyanov
- Publisher: AV Akademikerverlag
- ISBN-10: 3639402308
- ISBN-13: 9783639402308
- Format: 15.2 x 22.9 x 0.5 cm, minkšti viršeliai
- Language: English
- SAVE -10% with code: EXTRA
Reviews
Description
This work presents a framework based on a self-learning genetic algorithm for discovering prediction patterns in financial time series. By modifying a complex mathematical algorithm for evolutionary optimization in a manner more suitable for financial time series, specifics typical to asset trading were taken into account and were reflected in the solution set. In-sample the genetic algorithm was able to successfully find prediction rules allowing for the creation of successful, feasible and robust trading strategies with high performance. In order to justify the in-sample performance, the prediction patterns were tested on an out-of -sample time period: the results confirmed the ability of the genetic algorithm to find prediction patterns with robust performance. The proposed framework shows great potential when working with high dimensional search spaces and in finding non-linear dependencies in financial data.
EXTRA 10 % discount with code: EXTRA
The promotion ends in 21d.17:26:47
The discount code is valid when purchasing from 10 €. Discounts do not stack.
- Author: Antonio Stoyanov
- Publisher: AV Akademikerverlag
- ISBN-10: 3639402308
- ISBN-13: 9783639402308
- Format: 15.2 x 22.9 x 0.5 cm, minkšti viršeliai
- Language: English English
This work presents a framework based on a self-learning genetic algorithm for discovering prediction patterns in financial time series. By modifying a complex mathematical algorithm for evolutionary optimization in a manner more suitable for financial time series, specifics typical to asset trading were taken into account and were reflected in the solution set. In-sample the genetic algorithm was able to successfully find prediction rules allowing for the creation of successful, feasible and robust trading strategies with high performance. In order to justify the in-sample performance, the prediction patterns were tested on an out-of -sample time period: the results confirmed the ability of the genetic algorithm to find prediction patterns with robust performance. The proposed framework shows great potential when working with high dimensional search spaces and in finding non-linear dependencies in financial data.
Reviews