为什么是nan，type却不是np.nan或者np.NAN,np.Nan？

想计算mean结果却是nan，type又不是nan ，求解

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
gd=pd.read_csv('LBMA-GOLD.csv')
mk=pd.read_csv('BCHAIN-MKPRU.csv')

gd.sort_index(axis=1)
mk.sort_index(axis=1)
x1=gd['Date']
y1=gd['USD (PM)']
x2=mk['Date']
y2=mk['Value']

print(np.any(pd.isnull(y1)))
print(y1[74])
print(y1[74]==np.nan)

y1[74]是字符串的nan吗