2
$\begingroup$

US CPI monthly data

Hi, I'm trying to predict US inflation rate. The unit is in percentage change from a year ago. Would it be possible to use Fourier transform on the independent data to create a new feature, knowing that it's non stationary? If possible can I also get more reference? I'm new to time series, but did a course on forecasting.

Improve this question
$\endgroup$
2
  • 2
    $\begingroup$ You can compute anything you want, including the FT. The better question to ponder is whether that will be useful or meaningful to you. For predicting any inflation rate it's likely a bad procedure because inflation is not going to exhibit predictable perfect cyclic behavior, not even approximately. (That's an economic opinion, not a statistical one.) $\endgroup$
    – whuber ♦
    Commented Sep 28 at 14:52
  • 1
    $\begingroup$ this is what I thought also, glad my intuition wasn't that off. Thanks! $\endgroup$
    – Briefbreaddd
    Commented Sep 28 at 17:13

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.