I have some stock 5min data, like as:
Date Open High Low Close Volume
0 2024-11-19 09:35:00 11.75 11.79 11.55 11.78 32673600
1 2024-11-19 09:40:00 11.78 11.81 11.73 11.79 14802700
2 2024-11-19 09:45:00 11.79 11.84 11.79 11.82 13837400
3 2024-11-19 09:50:00 11.81 11.83 11.76 11.82 8534200
4 2024-11-19 09:55:00 11.82 11.87 11.80 11.87 8540500
5 2024-11-19 10:00:00 11.87 11.96 11.87 11.90 20659800
6 2024-11-19 10:05:00 11.89 11.90 11.82 11.82 11691000
7 2024-11-19 10:10:00 11.82 11.82 11.73 11.74 8762900
8 2024-11-19 10:15:00 11.74 11.74 11.71 11.73 6870500
9 2024-11-19 10:20:00 11.73 11.73 11.68 11.70 6244800
10 2024-11-19 10:25:00 11.70 11.70 11.66 11.69 5083000
11 2024-11-19 10:30:00 11.70 11.73 11.69 11.71 5342400
12 2024-11-19 10:35:00 11.72 11.74 11.71 11.73 3311800
13 2024-11-19 10:40:00 11.73 11.74 11.71 11.72 2331900
14 2024-11-19 10:45:00 11.72 11.72 11.70 11.72 3024100
15 2024-11-19 10:50:00 11.71 11.74 11.70 11.71 2774200
16 2024-11-19 10:55:00 11.70 11.72 11.70 11.71 1313000
17 2024-11-19 11:00:00 11.72 11.75 11.71 11.74 1737400
18 2024-11-19 11:05:00 11.75 11.75 11.73 11.75 1690600
19 2024-11-19 11:10:00 11.74 11.76 11.73 11.76 1751800
20 2024-11-19 11:15:00 11.76 11.76 11.72 11.73 2248700
21 2024-11-19 11:20:00 11.73 11.73 11.70 11.71 2464200
22 2024-11-19 11:25:00 11.71 11.71 11.69 11.70 1033600
23 2024-11-19 11:30:00 11.69 11.70 11.67 11.69 2063600
I use df.resample to convert them to 30m data, the code is:
df = df.set_index('Date')
df = df.resample('30T').agg({'Open':'first', 'High':'max', 'Low':'min','Close':'last',
'Volume':'sum'}, closed='right', label = 'right').dropna()
But I got strange result like these:
Open High Low Close Volume
Date
2024-11-19 09:30:00 11.75 11.87 11.55 11.87 78388400
2024-11-19 10:00:00 11.87 11.96 11.66 11.69 59312000
2024-11-19 10:30:00 11.70 11.74 11.69 11.71 18097400
2024-11-19 11:00:00 11.72 11.76 11.69 11.70 10926300
2024-11-19 11:30:00 11.69 11.70 11.67 11.69 2063600
Here are correct 30m data export from my trading software:
Time Open High Low Close Volume
2024/11/19-10:00 11.75 11.96 11.55 11.9 99048200
2024/11/19-10:30 11.89 11.9 11.66 11.71 43994600
2024/11/19-11:00 11.72 11.75 11.7 11.74 14492400
2024/11/19-11:30 11.75 11.76 11.67 11.69 11252500
The data at 9:30 is irrelevant, mainly because the following data are not correct. but I did not find more parameters of df.sample. How can I correctly aggregate the data?
I have some stock 5min data, like as:
Date Open High Low Close Volume
0 2024-11-19 09:35:00 11.75 11.79 11.55 11.78 32673600
1 2024-11-19 09:40:00 11.78 11.81 11.73 11.79 14802700
2 2024-11-19 09:45:00 11.79 11.84 11.79 11.82 13837400
3 2024-11-19 09:50:00 11.81 11.83 11.76 11.82 8534200
4 2024-11-19 09:55:00 11.82 11.87 11.80 11.87 8540500
5 2024-11-19 10:00:00 11.87 11.96 11.87 11.90 20659800
6 2024-11-19 10:05:00 11.89 11.90 11.82 11.82 11691000
7 2024-11-19 10:10:00 11.82 11.82 11.73 11.74 8762900
8 2024-11-19 10:15:00 11.74 11.74 11.71 11.73 6870500
9 2024-11-19 10:20:00 11.73 11.73 11.68 11.70 6244800
10 2024-11-19 10:25:00 11.70 11.70 11.66 11.69 5083000
11 2024-11-19 10:30:00 11.70 11.73 11.69 11.71 5342400
12 2024-11-19 10:35:00 11.72 11.74 11.71 11.73 3311800
13 2024-11-19 10:40:00 11.73 11.74 11.71 11.72 2331900
14 2024-11-19 10:45:00 11.72 11.72 11.70 11.72 3024100
15 2024-11-19 10:50:00 11.71 11.74 11.70 11.71 2774200
16 2024-11-19 10:55:00 11.70 11.72 11.70 11.71 1313000
17 2024-11-19 11:00:00 11.72 11.75 11.71 11.74 1737400
18 2024-11-19 11:05:00 11.75 11.75 11.73 11.75 1690600
19 2024-11-19 11:10:00 11.74 11.76 11.73 11.76 1751800
20 2024-11-19 11:15:00 11.76 11.76 11.72 11.73 2248700
21 2024-11-19 11:20:00 11.73 11.73 11.70 11.71 2464200
22 2024-11-19 11:25:00 11.71 11.71 11.69 11.70 1033600
23 2024-11-19 11:30:00 11.69 11.70 11.67 11.69 2063600
I use df.resample to convert them to 30m data, the code is:
df = df.set_index('Date')
df = df.resample('30T').agg({'Open':'first', 'High':'max', 'Low':'min','Close':'last',
'Volume':'sum'}, closed='right', label = 'right').dropna()
But I got strange result like these:
Open High Low Close Volume
Date
2024-11-19 09:30:00 11.75 11.87 11.55 11.87 78388400
2024-11-19 10:00:00 11.87 11.96 11.66 11.69 59312000
2024-11-19 10:30:00 11.70 11.74 11.69 11.71 18097400
2024-11-19 11:00:00 11.72 11.76 11.69 11.70 10926300
2024-11-19 11:30:00 11.69 11.70 11.67 11.69 2063600
Here are correct 30m data export from my trading software:
Time Open High Low Close Volume
2024/11/19-10:00 11.75 11.96 11.55 11.9 99048200
2024/11/19-10:30 11.89 11.9 11.66 11.71 43994600
2024/11/19-11:00 11.72 11.75 11.7 11.74 14492400
2024/11/19-11:30 11.75 11.76 11.67 11.69 11252500
The data at 9:30 is irrelevant, mainly because the following data are not correct. but I did not find more parameters of df.sample. How can I correctly aggregate the data?
Share Improve this question asked Nov 19, 2024 at 12:19 Sun JarSun Jar 3411 silver badge12 bronze badges2 Answers
Reset to default 1By default the reference in resample is the start of the day. It looks like you want the start of the data. You should set origin='start' instead of the default origin='start_day':
(df.resample('30min', origin='start')
.agg({'Open':'first', 'High':'max', 'Low':'min','Close':'last',
'Volume':'sum'}, closed='right', label = 'right')
.dropna()
)
Output:
Open High Low Close Volume
Date
2024-11-19 09:35:00 11.75 11.96 11.55 11.90 99048200
2024-11-19 10:05:00 11.89 11.90 11.66 11.71 43994600
2024-11-19 10:35:00 11.72 11.75 11.70 11.74 14492400
2024-11-19 11:05:00 11.75 11.76 11.67 11.69 11252500
And presumably, you want to pass the label parameter to resample:
(df.resample('30min', origin='start', label='right')
.agg({'Open':'first', 'High':'max', 'Low':'min','Close':'last',
'Volume':'sum'})
.dropna()
)
Output:
Open High Low Close Volume
Date
2024-11-19 10:05:00 11.75 11.96 11.55 11.90 99048200
2024-11-19 10:35:00 11.89 11.90 11.66 11.71 43994600
2024-11-19 11:05:00 11.72 11.75 11.70 11.74 14492400
2024-11-19 11:35:00 11.75 11.76 11.67 11.69 11252500
Finally, if you want to round to 30 minutes:
(df.resample('30min', origin='start', label='right')
.agg({'Open':'first', 'High':'max', 'Low':'min',
'Close':'last', 'Volume':'sum'})
.dropna()
.pipe(lambda x: x.set_axis(x.index.floor('30min')))
)
Output:
Open High Low Close Volume
Date
2024-11-19 10:00:00 11.75 11.96 11.55 11.90 99048200
2024-11-19 10:30:00 11.89 11.90 11.66 11.71 43994600
2024-11-19 11:00:00 11.72 11.75 11.70 11.74 14492400
2024-11-19 11:30:00 11.75 11.76 11.67 11.69 11252500
The closed='right' and label='right' parameters causing the intervals to align differently.
closed='right': This makes the intervals include the right endpoint but exclude the left.
label='right': This labels the resulting bins by the right edge of the interval.
df = pd.DataFrame(data)
df['Date'] = pd.to_datetime(df['Date'])
df.set_index('Date', inplace=True)
# Resampling
resampled_df = df.resample('30T', closed='left', label='right').agg({
'Open': 'first',
'High': 'max',
'Low': 'min',
'Close': 'last',
'Volume': 'sum'
}).dropna()
print(resampled_df)
1.Use closed='left' to include the left edge of each interval.
2.Use label='right' to match your trading software's labeling.
3.Use .agg() to compute the required statistics.