時系列 model
線形 model
AR (自己囘歸。autoregressive) model$ {\rm AR}(p)
$ X_t=c+\sum_{i=1}^p\varphi_i X_{t-i}+\varepsilon_t
$ p : model の次數 (order)
$ \varphi_i: model の parameter
$ c: 定數
線形反饋 (linear feed back)
MA (移動平均。moving average) model$ {\rm MA}(q)
$ X_t=c+\sum_{i=0}^q\theta_i\varepsilon_{t-i}
$ q: model の次數
$ \theta_i: model の parameter
$ c: 定數
線形前饋 (linear feed forward)
ARMA (自己囘歸移動平均。autoregressive moving average) model$ {\rm ARMA}(p,q)
$ X_t=c+\sum_{i=1}^p\varphi_i X_{t-i}+\sum_{i=0}^q\theta_i\varepsilon_{t-i}
ARIMA (自己囘歸和分移動平均。autoregressive integrated moving average) model$ {\rm ARIMA}(p,d,q)
SARIMA (季節變動自己囘歸和分移動平均。seasonal autoregressive integrated moving average) model
Prophet
狀態空閒 model
Kalman filter
粒子 filter (particle filter。逐次 Monte-Carlo 法 (sequential Monte-Carlo mothod。SMC))
ARCH (分散自己囘歸。分散不均一。autoregressive conditional heteroscedasticity) model$ {\rm ARCH}(q)
GARCH (一般化分散自己囘歸。generalized autoregressive conditional heteroscedasticity) model$ {\rm GARCH}(p,q)
後退作用素 (遲れ作用素。lag operator)
Granger 因果性