I usually see the $log$ transformation of prices: $$p_{new}\left(t\right) = ln\left(\frac{p_t}{p_{t-1}}\right), t \in [2...N]$$.

Let's our series be a trend stationary time series like: $$p\left(t\right) = kt + b + \xi(t)$$, where $k,b$ are numbers, $t \in [1...N]$, $\xi(t)$ is the random variable like $\xi(t) \sim N\left(\mu, \sigma\right)$.

For large $b$ and small $k, \sigma$ we have "good" transformed series, but if $b$ small and $\sigma$ big, so, we have "bad" transformed series.

"Good" ($k = 2, b = 100, \sigma = 3, t \in \left[0...100\right]$). "Good" origin series

"Good" transformed series

"Bad" ($k = 2, b = 10, \sigma = 10$).

"Bad" origin series

"Bad" transformed series

So, what's the correct method for TS-series transformation (econometric-style transformation)?

Thank you.

  • 2
    $\begingroup$ it's not entirely clear what you're trying to do and/or what your objective is. typically we transform one series or set of values to another because it makes them easier to work with or allows us to apply a certain set of techniques to the result. you wouldn't need to do that if you're simply simulating a price time-series with some set of parameters. it's also not clear what framework you're using to establish one TS as 'good' and the other 'bad' please consider re-working your question to make it clearer. $\endgroup$ – Chris May 7 at 17:18
  • $\begingroup$ Okay, I'll try it today. Thnxs for response. $\endgroup$ – Dmitriy May 8 at 12:03

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.