I want to generate a random data with variable rate of increment not fixed one so that it resemble to real world.

Thanks in advance

I am not able to get a distribution like above

CodePudding user response：

Generating concise data on such a sensitive topic is not easy. I answer this question with a purely demonstrative example, but I would refer you to reading scientific publications that explain well the underlying statistical criteria behind such studies.

A good approach would be to train a machine learning model on real pandemic data and then have it produce time-series forecasting.

If we want to construct a toy dataset, based on what we have been able to learn about the covid, we can make it resemble a Gaussian trend with arbitrary peaks. We can use

It may make sense to intervene on the parameters I put in by hand to smooth out the Gaussians a bit more and get a more sinus-like pattern. But again, this is a completely toy dataset.