Option to specify correlation type when calling df.pcorr()

[ozika]

Hi!
I think it would be useful to be able to specify the correlation type when calling pcorr() directly as pandas method.

I suppose the only change would be this:
On line 952:

V = self.cov()  # Covariance matrix

to

 if method == "spearman":
    # Convert the data to rank, similar to R cov()
    V = data.rank(na_option="keep").cov()
else:
    V = data.cov()

[raphaelvallat]

Hi @ozika,

Yes — please feel free to work on a PR (see contributing guidelines here)!

Thanks,
Raphael