4

I'm fitting a regression model $y_t$ to a time series $x_t$ (not a dynamic model involving ARMA terms!). I saw that useful predictors to put in my model are $t$, seasonality variables and lagged values of other time-series that are affecting my time-series.

question: why don't they mention the time series lagged variables themselves $x_{t-1}$, $x_{t-2}$? this is different than a dynamic regression model that uses $y_{t-1}$ or $\epsilon_{t-1}$, but sounds so simple and useful...

Richard Hardy
  • 54,375
  • 10
  • 95
  • 219
ihadanny
  • 2,596
  • 3
  • 19
  • 31
  • if you use a lagged variable of $y_t$ itself then that often has the same effect as including lagged $x_t$'s. Check out lagged dependent variable models or autoregressive distributed lag models for more details. – mlofton Nov 13 '19 at 15:09
  • what you need to do is to investigate transfer function models (a.k.a dynamic regression) and perhaps read https://stats.stackexchange.com/questions/221072/why-is-prewhitening-important/305634#305634 – IrishStat Nov 13 '19 at 22:09

0 Answers0