MemDA: Forecasting Urban Time Series with Memory-based Drift Adaptation.
Zekun CaiRenhe JiangXinyu YangZhaonan WangDiansheng GuoHill Hiroki KobayashiXuan SongRyosuke ShibasakiPublished in: CIKM (2023)
Keyphrases
- forecasting accuracy
- weather forecasting
- financial time series
- logistics demand
- arma model
- exponential smoothing
- box jenkins
- neural network model
- dynamic time warping
- hybrid model
- turning points
- short term
- forecasting model
- moving average
- support vector regression
- chaotic time series
- non stationary
- exchange rate
- bp neural network
- mackey glass
- urban planning
- stock market
- urban areas
- arima model
- concept drift
- multivariate time series
- memory based learning
- demand forecasting
- stock price
- garch model
- public transport
- autoregressive model
- regression model
- long term
- adaptation strategies
- prediction model
- phase space reconstruction
- grey model
- medium term