# Correlation function returning nan in pandas

**URL:** <https://forum.anaconda.com/t/correlation-function-returning-nan-in-pandas/40265>\
**Category:** Data Science, ML & AI\
**Created:** [September 27, 2022, 8:36pm UTC](https://forum.anaconda.com/t/correlation-function-returning-nan-in-pandas/40265 "2022-09-27T20:36:14Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![quique123](https://dub1.discourse-cdn.com/anaconda/user_avatar/forum.anaconda.com/quique123/32/1033_2.png) [@quique123](https://forum.anaconda.com/u/quique123)\
**Post date:** [September 27, 2022, 8:36pm UTC](https://forum.anaconda.com/t/correlation-function-returning-nan-in-pandas/40265/1 "2022-09-27T20:36:14Z")

</div>

Im running this code to compare these 2 series but the correlation function returns nan:

```auto
import pandas as pd
from pandas import Series
%matplotlib inline
from statsmodels.tsa.seasonal import seasonal_decompose
from sklearn.preprocessing import RobustScaler

# Seasonal exploration visualizing ts
import matplotlib.pyplot as plt
import seaborn as sns
import matplotlib.dates as mdates

# Set of commands to plot 1 column from multiple series on excel file
df=pd.read_excel('DataLSTMReady.xlsx')
df.Date=pd.to_datetime(df.Date)

df=df.set_index('Date')

df=df.iloc[:,[0,1,3]] #df=df.iloc[:,0:4]

print(df)

# COMPUTE CORRELATION BETWEEN TIMESERIES USING pctchange INSTEAD OF LEVELS/VALUES OF EACH
df['107057_Ret'] = df['107057'].pct_change()
df['Infl_Mens_Ret']= df['Infl Mens'].pct_change()
df['107057-mkp_Ret']= df['107057-mkp'].pct_change()

plt.scatter(df['107057_Ret'],df['107057-mkp_Ret'])
plt.show

correlation=df['107057_Ret'].corr(df['107057-mkp_Ret'])
print("Correlation is: ", correlation)

sns.set(rc={'figure.figsize':(11, 4)})

```

Here are the results:

 ![Screen Shot 2022-09-27 at 2.35.47 PM](https://europe1.discourse-cdn.com/anaconda/original/2X/d/d95359b30024e1b9a06576bf8f13a903b930f4fb.png)

---

<div class="post-metadata">

**Author:** ![quique123](https://dub1.discourse-cdn.com/anaconda/user_avatar/forum.anaconda.com/quique123/32/1033_2.png) [@quique123](https://forum.anaconda.com/u/quique123)\
**Post date:** [September 27, 2022, 11:42pm UTC](https://forum.anaconda.com/t/correlation-function-returning-nan-in-pandas/40265/2 "2022-09-27T23:42:00Z")

</div>

ok i had some zero values in my series…correlation comes to -0.05 anyway…lol

of course i saw it on the graph…