ewm() for exponential moving average in Pandas - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

Output of ewm() with span parameter

What is the output of the following code snippet?

Pandas

import pandas as pd
s = pd.Series([10, 20, 30, 40, 50])
ewm_result = s.ewm(span=3, adjust=False).mean()
print(ewm_result.round(2))

A [10.00, 15.00, 22.50, 31.25, 40.62]

B [10.00, 13.33, 19.44, 27.78, 37.96]

C [10.00, 20.00, 30.00, 40.00, 50.00]

D [10.00, 16.67, 25.56, 35.37, 45.91]

💡 Hint

Remember that adjust=False means the weights are calculated recursively without adjusting for prior weights.

With span=3 and adjust=False, the ewm() calculates the exponential moving average giving more weight to recent values. The output is a series of smoothed values starting at the first value.

❓ data_output

intermediate

Number of items in ewm result with min_periods

How many non-NaN values are in the result of this code?

Pandas

import pandas as pd
s = pd.Series([5, 10, 15, 20, 25])
ewm_result = s.ewm(span=2, min_periods=3).mean()
print(ewm_result.dropna().count())

A 3

B 2

C 5

D 0

💡 Hint

min_periods sets the minimum number of observations required to have a value.

Since min_periods=3, the first two values are NaN. From the third value onward, the ewm mean is calculated, so 3 values are non-NaN.

🔧 Debug

advanced

Identify the error in ewm() usage

What error does this code raise?

Pandas

import pandas as pd
s = pd.Series([1, 2, 3, 4])
ewm_result = s.ewm(com=0, span=2).mean()

A No error, runs successfully

B TypeError: ewm() got multiple values for argument 'span'

C ValueError: com and span cannot be specified at the same time

D AttributeError: 'Series' object has no attribute 'ewm'

💡 Hint

Check the parameters passed to ewm().

You cannot specify both 'com' and 'span' parameters together in ewm(). This raises a ValueError.

🚀 Application

advanced

Using ewm() to smooth noisy data

You have a noisy time series data. Which ewm() parameter setting will give the smoothest result?

A halflife=1, adjust=True

B span=10, adjust=True

C span=2, adjust=False

D com=0.5, adjust=False

💡 Hint

A larger span means more smoothing.

A larger span (10) with adjust=True gives more weight to all past data, resulting in smoother output.

🧠 Conceptual

expert

Effect of adjust parameter in ewm()

Which statement best describes the effect of adjust=False in pandas ewm()?

A Weights are normalized so the sum equals 1, considering all past data.

B The ewm() function raises an error if adjust=False is used.

C The result is a simple moving average ignoring weights.

D Weights are calculated recursively without normalization, giving more weight to recent data.

💡 Hint

Think about how weights are applied when adjust=False.

With adjust=False, the calculation uses recursive weights without normalizing over all past data, emphasizing recent points more.