A Hybrid Time Series Model based on Dilated Conv1D and LSTM with Applications to PM2.5 Forecasting.
Liqing ZhaoBo ChengJunliang ChenPublished in: Aust. J. Intell. Inf. Process. Syst. (2019)
Keyphrases
- weather forecasting
- forecasting accuracy
- financial time series
- chaotic time series
- short term
- recurrent neural networks
- arma model
- exponential smoothing
- box jenkins
- multivariate time series
- non stationary
- hybrid model
- stock market
- garch model
- long term
- mackey glass
- forecasting model
- dynamic time warping
- subsequence matching
- forecast models
- demand forecasting
- medium term
- reinforcement learning
- grey model
- elman neural network
- data sets
- electricity consumption
- bp neural network
- hybrid approaches
- exchange rate
- hybrid learning
- symbolic representation
- autoregressive