我已经编写了这段代码,但是当我尝试使用Series函数
import pandas as pd
import numpy as np
import datetime as dt
# example values
dates = np.asarray(pd.date_range('2000-01-01', periods=200))
# create a dataframe
df = pd.DataFrame(np.random.randn(200, 4), index=dates, columns=['A', 'B', 'C', 'D'])
print(df)
start_idx = np.where(dates>=np.datetime64(dt.datetime.strptime('2000-01-01', "%Y-%m-%d")))[0][-1]
dateIndex = df[0: start_idx+1].index
print(dateIndex)
historydict = {}
for i in range(3):
returnHistory = pd.Series(df[i+1:start_idx+2]['A'])
returnHistory.index = dateIndex
# the above two periods assigned to date t, the returns lagged by i periods
rname="Hist"+str(i+1)
historydict[rname]=returnHistory
hDataFrame = pd.DataFrame(historydict)
print(hDataFrame)
historydict将用于跟踪t-1,t-2,t-3的序列,每个序列将成为数据帧中的一列,但
returnHistory.index = dateIndex
给出此错误:
ValueError: Length mismatch: Expected axis has 199 elements, new values have 200 elements
答案 0 :(得分:1)
在线
returnHistory = pd.Series(temp)
是形状为pandas.DataFrame
的{{1}}对象。试试
(199, 4)
或其他列。