Pandas中的日期时间date处理小结

import pandas as pd

# 将字符串转换为datetime
date_str = ['2023-01-01', '2023-01-02', '2023-01-03']
dates = pd.to_datetime(date_str)
print(dates)

# 创建日期范围
date_range = pd.date_range('2023-01-01', periods=5, freq='D')
print(date_range)

# 带有时区的日期范围
date_range_tz = pd.date_range('2023-01-01', periods=5, freq='D', tz='Asia/Shanghai')
print(date_range_tz)

# 创建示例DataFrame
df = pd.DataFrame({
    'date': pd.date_range('2023-01-01', periods=5, freq='D'),
    'value': [10, 20, 30, 40, 50]
})

# 提取年、月、日等属性
df['year'] = df['date'].dt.year
df['month'] = df['date'].dt.month
df['day'] = df['date'].dt.day
df['day_of_week'] = df['date'].dt.dayofweek  # 周一=0, 周日=6
df['day_name'] = df['date'].dt.day_name()
df['is_weekend'] = df['date'].dt.dayofweek >= 5

print(df)

# 计算时间差
df['date_diff'] = df['date'] - df['date'].shift(1)
print(df[['date', 'date_diff']])

# 使用Timedelta进行时间运算
df['date_plus_2days'] = df['date'] + pd.Timedelta(days=2)
df['date_plus_3hours'] = df['date'] + pd.Timedelta(hours=3)
print(df)

# 日期比较
start_date = pd.to_datetime('2023-01-02')
df['after_start_date'] = df['date'] > start_date
print(df[['date', 'after_start_date']])

# 创建示例时间序列数据
ts = pd.Series(
    [1, 2, 3, 4, 5],
    index=pd.date_range('2023-01-01', periods=5, freq='D')
)

# 降采样（低频） - 计算每周平均值
weekly = ts.resample('W').mean()
print("Weekly resample:\n", weekly)

# 升采样（高频） - 填充缺失值
hourly = ts.resample('H').ffill()
print("Hourly resample (forward fill):\n", hourly.head(10))  # 只显示前10行

# 本地化时区
ts = pd.Series(
    [1, 2, 3],
    index=pd.date_range('2023-01-01', periods=3, freq='D')
)
ts = ts.tz_localize('UTC')
print("UTC timezone:\n", ts)

# 时区转换
ts_shanghai = ts.tz_convert('Asia/Shanghai')
print("Shanghai timezone:\n", ts_shanghai)

# 设置日期为索引
df.set_index('date', inplace=True)

# 按年份切片
print(df.loc['2023'])

# 按月份切片
print(df.loc['2023-01'])

# 按日期范围切片
print(df.loc['2023-01-02':'2023-01-04'])

# 读取包含日期时间的数据
# 假设有一个CSV文件包含日期列
# df = pd.read_csv('data.csv', parse_dates=['date_column'])

# 处理缺失日期
full_date_range = pd.date_range(start=df.index.min(), end=df.index.max(), freq='D')
df = df.reindex(full_date_range)

# 填充缺失值
df['value'] = df['value'].fillna(method='ffill')  # 前向填充

# 计算滚动平均值
df['7_day_avg'] = df['value'].rolling(window='7D').mean()

print(df.head(10))

from pandas.tseries.offsets import CustomBusinessDay
from pandas.tseries.holiday import USFederalHolidayCalendar

# 使用美国联邦假日日历
us_bd = CustomBusinessDay(calendar=USFederalHolidayCalendar())
date_range = pd.date_range('2023-01-01', periods=10, freq=us_bd)
print("US business days only:\n", date_range)

# 创建季度数据
quarterly = pd.Series(
    [100, 200, 300, 400],
    index=pd.date_range('2023-01-01', periods=4, freq='Q')
)

# 季度开始和结束日期
print("Quarter start:\n", quarterly.index)
print("Quarter end:\n", quarterly.index + pd.offsets.QuarterEnd())