I have a time series $Y_t$ (example:university applications received in a certain month) which I want to forecast. I have another time series $X_t$ and I know that $Y_t$ is related to past lags of $X_t$ (example: high school graduates per month), meaning that a certain amount of $X_t$ at month $t$ will determine the amount of $Y_t$, say, 12 months later. I also know that the probability applying at the university is decreasing with every month once graduated. A lot of students will apply right after graduating but only a small percentage will apply, say, 6 month later. $X_t$ has a seasonal pattern and $Y_t$ has several shocks because of law changes.

My questions are: Is a dynamic regression model/ARIMAX appropriate to forecast $Y_t$ with $X_{t-k}$? How do I deal with the shocks in $Y_t$ knowing that they hide the underlying structure of this time series?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.