模型估计为ARIMA(4,0,2)，即ARMA(4,2)
系数为：
ar1 ar2 ar3 ar4 ma1 ma2
-0.5505 0.2316 0.0880 -0.4325 -0.1944 -0.5977
s.e. 0.1657 0.1428 0.1402 0.1270 0.1766 0.1732
s.e.是系数的标准差，系数显著性要自己算，|系数/se| >1.96 即 95％的置信度
sigma^2 estimated 估计值方差
log likelihood 对数似然值
（这个不用解释了吧）
AIC=709.13 AICc=710.73 BIC=725.63
再就是下面一堆误差计算
ME Mean Error
RMSE Root Mean Squared Error
MAE Mean Absolute Error
MPE Mean Percentage Error
MAPE Mean Absolute Percentage
MASE Mean Absolute Scaled Error
stats包Description
Use Kalman Filtering to find the (Gaussian) log-likelihood, or for
forecasting or smoothing.
Usage
KalmanLike(y, mod, nit = 0, fast = TRUE)
KalmanRun(y, mod, nit = 0, fast = TRUE)
KalmanSmooth(y, mod, nit = 0)
KalmanForecast(n.ahead = 10, mod, fast = TRUE)
makeARIMA(phi, theta, Delta, kappa = 1e6)
