AR time series analysis

Too few periods for decompose() [closed]

Your time series, however you created it, has no seasonal cycles or less than 2 seasonal cycles. (This may not indicate that the data are not seasonal; possibly you created samplets incorrectly.) 

1.white.test

white.test(x, lag = 1, qstar = 2, q = 10, range = 4, type = c("Chisq","F"), scale = TRUE, …)

2.seasons explainations

https://www.spscommerce.com/blog/shopping-seasons-consumer-spsa/

3.自相关和偏自相关

参考 https://blog.youkuaiyun.com/yujunbeta/article/details/8091411

偏自相关就是在试图解释在剔除了中间k-1个随机变量x(t-1)、x(t-2)、……、x(t-k+1)的干扰之后，x(t-k)对x(t)影响的相关程度。PACF

ACF starts at lag 0 and PACF starts at lag 1.

4. AR 和 MA 拖尾性和结尾性

AR（2）模型的偏自相关函数是截尾的（但由于这个是数据，所以出现pacf只能看出趋势，而不是在2步后直接变为0）

R 模型实现 MA《

https://zhuanlan.zhihu.com/p/26310640

5. Applied timeSeries 教程， 滨州大学

https://onlinecourses.science.psu.edu/stat510/node/62/

6 with seasonal 

https://stats.stackexchange.com/questions/223316/modelling-moving-holiday-effects-in-forecasting

use dummy variables for the holiday

论文

https://www.census.gov/srd/papers/pdf/RRS2018-01.pdf

ARMA 实现 ，并上 holiday

https://stackoverflow.com/questions/46873899/weekly-forecasts-with-holidays

7. 教科书式说明

https://a-little-book-of-r-for-time-series.readthedocs.io/en/latest/src/timeseries.html#decomposing-time-series

8. nonstationary time series

adf.test()

https://stats.stackexchange.com/questions/27332/how-to-know-if-a-time-series-is-stationary-or-non-stationary

9.