Question

Suppose that we didn't drop the missing incomes. What will running df['income'].mean() output

Please answer question 2C and 2D.

Thank you!

Question 2c: Note that we can now compute the average income. If your df variable contains the right values, dfl'income' ].mean the value 25474.07 should produce Suppose that we didn't drop the missing incomes. What will running dfI'income'.mean ) output? Use the variable q2c_answer to record which of the below statements you think is true. As usual, you can use the cel you set q2c_answer once you've finished l below to run any code you' d like in order to help you answer this question as long as 1. dfl 'income' ].mean) will produce an error 2. df[ 'income'].mean) will output 0 3. dfl 'income' ].mean() will output nan (not a number) 4. No change; df'income' ].mean wll ignore the missing values and output 25474.07 5. dfl 'income'].mean) will fill in the missing values with the average income, then compute the average 6. df[ 'income' ] .mean) will fill in the missing values with 0, then compute the average In [ ]: | # YOUR CODE HERE In 1Tests for 2c assert isinstance (q2c_answer, int) Question 2d: Suppose that missing incomes did not occur at random, and that individuals with incomes below $10000 a year are less likely to report their incomes. If so, one of the statements is true. Record your choice in the variable q2d_answer. 1. df[ 'income' ].mean) will likely output a value that is smaller than the population's average income 2. df 'income' ] .mean) will likely output a value that is larger than the population's average income 3. df[ 'income'].mean will likely output a value that is the same as the population's average income 4. df['income'].mean) will raise an error. In [ ]; # YOUR CODE HERE raise NotImplementedError() In [ ]: # Tests for 2d assert isinstance(q2d_answer, int)

Answer 1

in question 2c. option d is correct. the statement df['income'].mean() would just ignore the nan values and compute the mean with the remaining values.

for example, in the below picture, column A, B and D has some nan values but the mean is calculated using the remaining values.

In [15]: 1 import pandas as pd 2 df pd.DataFrame(["A": [17, 4, 5, None, 1], "B":(7, 2, 34, 3, None], "C": [20, 57, 11, 3, 8], "D" [1, 35, None, 2, 43])) 4 6 df.mean() Out[15]: A 5.5e В 16.50 C 11.6e D 6.25 dtype: float64