Is there any way to plot 95% CI around a numpy array in Python?
The closest I managed to find was 'seaborn.pointplot' but didn't really work out for me somehow
Shows an error :
AttributeError: 'numpy.ndarray' object has no attribute 'get'
You could maybe try statsmodel library
https://www.statsmodels.org/stable/generated/statsmodels.graphics.tsaplots.plot_acf.html
Sathyanarayan Rao : I was already aware of this one. It provides autocorrelation with confidence intervals specified by the alpha parameter. The statsmodels library provides ACF and PACF plots but not the cross-correlation plots.
The xcorr function for cross correlation plots is present in the matplotlib library, but the problem is that it does not have any provision for the confidence intervals.
Hi Utkarsh. I was having the same problem as you have and I thought I'd post my solution here for future users.
Matlab has a similar function, called crosscorr(). This does plot the confidence bounds. In this function they use +/- 2/sqrt(N) with N the shortest length of the arrays you want to cross correlate. This gives the 95% confidence interval. https://github.com/smmckib/crosscorr/blob/master/crosscorr.m
I suppose this is a good enough approximation. However there are other more precise, yet more difficult ways to do this as described in this paper:
https://journals.ametsoc.org/doi/pdf/10.1175/1520-0442%281997%29010%3C2147%3AAMTETS%3E2.0.CO%3B2
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Programming Languages
- Python