0
$\begingroup$

Consider the following types of financial time series for a single publicly-listed stock:

  1. Price data
  2. Log returns
  3. Cumulative returns

Each is computed from the item listed before it: log returns are based on differences of prices, and cumulative returns are cumulative products of log returns.

  • Which of the random variables listed above possess a probability distribution function (PDF),
  • which have a cumulative distribution function (CDF), and
  • which have both a PDF and CDF?
  • for what sort of financial applications is the CDF preferred over the PDF, and vice versa?

I ask because the following post says all random variables have a CDF, but not all of them have a PDF. So I wanted to see how this applies to commonly used financial data, which are prices and returns. Graphical depictions of the above datas' CDF and PDFs displayed side-by-side would help in the explanation.

I'm particularly curious about cumulative returns. Since they're cumulative, it automatically makes me think it corresponds and is represented best by a CDF, so in a way I'm wondering if cumulative returns are more useful than they're made out to be, despite being non-stationary.

$\endgroup$
1
  • 1
    $\begingroup$ They don't have pdf because they have a pmf. $\endgroup$
    – confused
    Jul 23, 2020 at 12:37

1 Answer 1

5
$\begingroup$

For a continuous variable the PDF is the derivative of CDF. So returns or prices don't have a pdf if the cdf is not differentiable, e.g. it "jumps" at some point. The simplest models we use, like normally distributed log-returns, imply that returns, cumulative returns and prices all have a pdf.

$\endgroup$
4
  • $\begingroup$ You say all 3 have a pdf. And which of them have cdf? Stock returns usually aren't normal log returns, stocks have nonnormal log returns. $\endgroup$
    – develarist
    Jul 20, 2020 at 8:58
  • 1
    $\begingroup$ @develarist As you said any random variable has a cdf. I just meant we commonly assume all 3 have a pdf. Most of the common distributions like normal, t, gamma all have a pdf., normality is not important here. But you can find counterexamples mentioned in the previous post like some mixture models for which the pdf does not exist. $\endgroup$
    – fes
    Jul 20, 2020 at 13:06
  • $\begingroup$ @develarist I also don't see why one of these would be more likely to have a pdf than the others. Typically if you assume returns don't have a pdf, then prices won't either. $\endgroup$
    – fes
    Jul 20, 2020 at 13:10
  • $\begingroup$ here is a separate question, quant.stackexchange.com/questions/55797 $\endgroup$
    – develarist
    Jul 20, 2020 at 13:24

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.