Resampling time series data in Pandas - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

What is the output of this resampling code?

Given a time series with daily data, what is the output after resampling to monthly frequency using sum aggregation?

Pandas

import pandas as pd
import numpy as np
rng = pd.date_range('2023-01-01', periods=5, freq='D')
data = pd.Series([1, 2, 3, 4, 5], index=rng)
result = data.resample('M').sum()
print(result)

A

2023-01-31    15
Freq: M, dtype: int64

B

2023-01-05    15
Freq: D, dtype: int64

C

2023-01-01    1
2023-01-02    2
2023-01-03    3
2023-01-04    4
2023-01-05    5
Freq: D, dtype: int64

D

2023-01-31    5
Freq: M, dtype: int64

💡 Hint

Think about what resampling to 'M' (month end) does and how sum aggregates daily values.

Resampling to 'M' groups all days in January and sums their values. The sum of 1+2+3+4+5 is 15, indexed at the last day of January.

❓ data_output

intermediate

How many rows are in the resampled DataFrame?

A DataFrame has hourly data for 2 days. After resampling to daily frequency using mean aggregation, how many rows does the result have?

Pandas

import pandas as pd
import numpy as np
rng = pd.date_range('2023-01-01', periods=48, freq='H')
data = pd.DataFrame({'value': np.arange(48)}, index=rng)
result = data.resample('D').mean()
print(len(result))

A 2

B 1

C 24

D 48

💡 Hint

Resampling groups data by day. How many days are in 48 hours?

48 hours cover exactly 2 days, so resampling by day results in 2 rows, one for each day.

🔧 Debug

advanced

What error does this resampling code raise?

Identify the error raised by this code snippet:

Pandas

import pandas as pd
import numpy as np
rng = pd.date_range('2023-01-01', periods=5, freq='D')
data = pd.Series(np.arange(5), index=rng)
result = data.resample('M').mean()
result = result.resample('H').sum()

A AttributeError: 'Series' object has no attribute 'resample'

B ValueError: Cannot resample from monthly to hourly frequency

C No error, code runs successfully

D TypeError: Only valid with DatetimeIndex, TimedeltaIndex or PeriodIndex, but got an instance of 'Float64Index'

💡 Hint

Check the index type after the first resample and if resample supports chaining.

The first resample returns a Series with a DatetimeIndex at month end. Resampling from monthly to hourly is allowed and works without error.

🚀 Application

advanced

Which option produces the correct weekly resampled sum?

Given daily sales data, which code correctly resamples to weekly sums with weeks starting on Monday?

Pandas

import pandas as pd
import numpy as np
rng = pd.date_range('2023-01-01', periods=14, freq='D')
data = pd.Series(np.arange(14), index=rng)

A data.resample('7D').sum()

B data.resample('W-MON').sum()

C data.resample('W-SUN').sum()

D data.resample('W').sum()

💡 Hint

Check the alias for weekly resampling starting on Monday.

'W-MON' means weekly frequency with weeks ending on Monday, so sums are grouped accordingly. 'W' defaults to Sunday week end.

🧠 Conceptual

expert

What is the effect of using 'label' and 'closed' parameters in resample?

In pandas resample, what do the 'label' and 'closed' parameters control?

A 'label' controls the index type after resampling; 'closed' controls whether missing data is filled

B 'label' controls the aggregation function; 'closed' controls the frequency of resampling

C 'label' controls the sorting order of the result; 'closed' controls the data type of the output

D 'label' controls the timestamp assigned to the resampled bin; 'closed' controls which side of the interval is included

💡 Hint

Think about how intervals are labeled and which endpoints are included in resampling bins.

'label' sets whether the bin is labeled by its start or end timestamp. 'closed' sets which endpoint (left or right) is included in the bin interval.