西西河

主题:【原创】谈谈时间序列的平稳性(1) -- 万里风中虎

共:💬217 🌺809
全看分页树展 · 主题 跟帖
家园 Questions

1. 中国股市是否有上升趋势,如果有,数据就是非平稳的

Trend->non-stationarity is not equivalent to non-stationarity->Any trend, not to mention you claim up trend.

2. ADF test fails to reject non-stationarity,but it also rejects time trend (I assume '_trend' is time trend t). Honestly, failing to reject unit root is not impressive at all-it's supposed to fail on any stock price.

3. Since price fails ADF test, I would be extremely cautious when running OLS as the results may be spurious. I don't know your definition of variable 'number'. But did you check non-stationarity of this series and cointegration between X and Y?

4. I think mean-reversion model is a better candidate for your investigation of this topic

全看分页树展 · 主题 跟帖


有趣有益,互惠互利;开阔视野,博采众长。
虚拟的网络,真实的人。天南地北客,相逢皆朋友

Copyright © cchere 西西河