Pandas profiling correlation coefficient value Different from Pandas .corr()

ydataai / ydata-profiling

1 Line of code data quality profiling & exploratory data analysis for Pandas and Spark DataFrames.

https://docs.profiling.ydata.ai

MIT License

12.52k stars 1.68k forks source link

Pandas profiling correlation coefficient value Different from Pandas .corr() #1242

Open anandece4u opened 1 year ago

anandece4u commented 1 year ago

With pandas .corr() print(f"pearson = ",combine_df_2_num_2['noOfNetCredit'].corr(combine_df_2_num_2['medianBalance'], method='pearson')) I am getting output "pearson = 0.2130989508716667"

But in pandas profile I am getting very high correlation for the same variable pair as 0.955.

pandas version 1.4.4 pandas_profiling version 3.2.0

I am not able to understand why this difference.

fabclmnt commented 1 year ago

@anandece4u can you please provide details on what correlation matrix you are checking the value for?