Based on the tensorflow and keras, I dig into the machine learning algorithm of time series prediction with LSTM and other related methods. I compare the performance of them and invent more interesting structures for better and efficient forecasting.
Stock market is non-linear, dynamic and complicated system. I manage to use the previous informations of certain stock price to predict a credible future trend. I am still working for this to reach much better performance and help investors to make more reliable decisions.
Aiming at establishing the accurate and adaptive model to estimate the state of charge and health of Li-ion battery, we use the multivariate time series prediction model to map the relationship between voltage, current, temperature and battery state of charge.
Housing Market and Pridictive Analytics.
Time Series Prediction Method, Financial Market Dynamics and Trading Algorithm.
Predictive Models of Battery State of Charge and Fast Charging Strategy.