Home | About us | Methodologies | Progress | Results | References |
Updated on 17/04/2019 | |
November 2018 | - The direction of n-th day was the same that of the n-1 day. - Average accuracy: 50% |
January 2019 |
- Bernoulli naive Bayes classifier was used to predict daily trend. - Indicators included: moving average, relative strength index, stochastic oscillator - Average accuracy: 52.60% |
February 2019 |
- Bernoulli naive Bayes classifier was used. - Number of days included in the calculations of the indicators was adjusted for a higher accuracy. |
April 2019 |
- Naive Bayes and Principal Component Analysis were compared. They achieved similar accuracy. - Parameters of some popular technical indicators were tested. Common parameters did not provide the best results. |