CODEWITHSUNDEEP
pythonpandas
Rtrader
Rtrader

Reputation: 947

Convert Cumulative Returns to Daily Returns using pandas

So I am trying to go from cumulative returns given by 

hist_data = (hist_data - hist_data.iloc[0]) / hist_data.iloc[0]

And I am trying to go from this cumulative return to daily returns but am blanking on how to do this effectively. Any ideas? I thought this might work if I subtract by one

(hist_data[1:] - hist_data[:-1]) / hist_data[:-1]

Upvotes: 1

Views: 929

Answers (1)

andrew
andrew

Reputation: 2574

If hist_data contains the cumulative returns, then this is a common shortcut for computing daily returns. Assuming hist_data is a vector of return percentages, you will need to add 1.0 to hist_data, as I have done below. This way we have a vector of return ratios instead of return percentages.

import numpy as np

daily_returns = np.exp(np.log(hist_data + 1.0).diff())

Here we are simply using the property of natural logs (ln) that says 

ln(a/b) = ln(a) - ln(b)
   a/b  = exp(ln(a) - ln(b))

Upvotes: 5

Related Questions