Answer: Among 2500 beds and outdoor furniture items, the proportion of beds.
Among 2500 beds and outdoor furniture items, the number of beds.

Our test statistic needs to be able to distinguish between the two hypotheses. The first option does not do this, because it includes an absolute value. If the absolute difference between the proportion of beds and the proportion of outdoor furniture were large, it could be because IKEA sells more beds than outdoor furniture, but it could also be because IKEA sells more outdoor furniture than beds.

The second option is a valid test statistic, because if the proportion of beds is large, that suggests that the alternative hypothesis may be true.

Similarly, the third option works because if the number of beds (out of 2500) is large, that suggests that the alternative hypothesis may be true.

The fourth option is invalid because out of 2500 beds and outdoor furniture items, the number of beds plus the number of outdoor furniture items is always 2500. So the value of this statistic is constant regardless of whether the alternative hypothesis is true, which means it does not help you distinguish between the two hypotheses.

Difficulty: ⭐️⭐️

The average score on this problem was 78%.

Answer: Reading the table from top to bottom, the five expressions should be used in the following blanks: None, (b), (a), (c), None.

The correct way to define obs_diff is

    outdoor = (app_data.get('category')=='outdoor') 
    bed = (app_data.get('category')=='bed')
    obs_diff = (app_data[outdoor].shape[0] - app_data[bed].shape[0]) / app_data[outdoor | bed].shape[0]

The first provided line of code defines a boolean Series called outdoor with a value of True corresponding to each outdoor furniture item in app_data. Using this as the condition in a query results in a DataFrame of outdoor furniture items, and using .shape[0] on this DataFrame gives the number of outdoor furniture items. So app_data[outdoor].shape[0] represents the number of outdoor furniture items in app_data. Similarly, app_data[bed].shape[0] represents the number of beds in app_data. Likewise, app_data[outdoor | bed].shape[0] represents the total number of outdoor furniture items and beds in app_data. Notice that we need to use an or condition (|) to get a DataFrame that contains both outdoor furniture and beds.

We are told that the test statistic should be the proportion of outdoor furniture minus the proportion of beds. Translating this directly into code, this means the test statistic should be calculated as

    obs_diff = app_data[outdoor].shape[0]/app_data[outdoor | bed].shape[0] - app_data[bed].shape[0]) / app_data[outdoor | bed].shape[0]

Since this is a difference of two fractions with the same denominator, we can equivalently subtract the numerators first, then divide by the common denominator, using the mathematical fact \frac{a}{c} - \frac{b}{c} = \frac{a-b}{c}.

This yields the answer

    obs_diff = (app_data[outdoor].shape[0] - app_data[bed].shape[0]) / app_data[outdoor | bed].shape[0]

Notice that this is the observed value of the test statistic because it’s based on the real-life data in the app_data DataFrame, not simulated data.

Difficulty: ⭐️

The average score on this problem was 90%.

Answer: Way 1, Way 3, Way 4, Way 6

Let’s consider each way in order.

Way 1 is a correct solution. This code begins by defining a variable multi which will evaluate to an array with two elements representing the number of items in each of the two categories, after 2500 items are drawn randomly from the two categories, with each category being equally likely. In this case, our categories are beds and outdoor furniture, and the null hypothesis says that each category is equally likely, so this describes our scenario accurately. We can interpret multi[0] as the number of outdoor furniture items and multi[1] as the number of beds when we draw 2500 of these items with equal probability. Using the same mathematical fact from the solution to Problem 8.2, we can calculate the difference in proportions as the difference in number divided by the total, so it is correct to calculate the test statistic as (multi[0] - multi[1])/2500.

Way 2 is an incorrect solution. Way 2 is based on a similar idea as Way 1, except it calls np.random.multinomial twice, which corresponds to two separate random processes of selecting 2500 items, each of which is equally likely to be a bed or an outdoor furniture item. However, is not guaranteed that the number of outdoor furniture items in the first random selection plus the number of beds in the second random selection totals 2500. Way 2 calculates the proportion of outdoor furniture items in one random selection minus the proportion of beds in another. What we want to do instead is calculate the difference between the proportion of outdoor furniture and beds in a single random draw.

Way 3 is a correct solution. Way 3 does the random selection of items in a different way, using np.random.choice. Way 3 creates a variable called choice which is an array of 2500 values. Each value is chosen from the list [0,1] with each of the two list elements being equally likely to be chosen. Of course, since we are choosing 2500 items from a list of size 2, we must allow replacements. We can interpret the elements of choice by thinking of each 1 as an outdoor furniture item and each 0 as a bed. By doing so, this random selection process matches up with the assumptions of the null hypothesis. Then the sum of the elements of choice represents the total number of outdoor furniture items, which the code saves as the variable choice_sum. Since there are 2500 beds and outdoor furniture items in total, 2500 - choice_sum represents the total number of beds. Therefore, the test statistic here is correctly calculated as the number of outdoor furniture items minus the number of beds, all divided by the total number of items, which is 2500.

Way 4 is a correct solution. Way 4 is similar to Way 3, except instead of using 0s and 1s, it uses the strings 'bed' and 'outdoor' in the choice array, so the interpretation is even more direct. Another difference is the way the number of beds and number of outdoor furniture items is calculated. It uses np.count_nonzero instead of sum, which wouldn’t make sense with strings. This solution calculates the proportion of outdoor furniture minus the proportion of beds directly.

Way 5 is an incorrect solution. As described in the solution to Problem 8.2, app_data[outdoor|bed] is a DataFrame containing just the outdoor furniture items and the beds from app_data. Based on the given information, we know app_data[outdoor|bed] has 2500 rows, 1000 of which correspond to beds and 1500 of which correspond to furniture items. This code defines a variable samp that comes from sampling this DataFrame 2500 times with replacement. This means that each row of samp is equally likely to be any of the 2500 rows of app_data[outdoor|bed]. The fraction of these rows that are beds is 1000/2500 = 2/5 and the fraction of these rows that are outdoor furniture items is 1500/2500 = 3/5. This means the random process of selecting rows randomly such that each row is equally likely does not make each item equally likely to be a bed or outdoor furniture item. Therefore, this approach does not align with the assumptions of the null hypothesis.

Way 6 is a correct solution. Way 6 essentially modifies Way 5 to make beds and outdoor furniture items equally likely to be selected in the random sample. As in Way 5, the code involves the DataFrame app_data[outdoor|bed] which contains 1000 beds and 1500 outdoor furniture items. Then this DataFrame is grouped by 'category' which results in a DataFrame indexed by 'category', which will have only two rows, since there are only two values of 'category', either 'outdoor' or 'bed'. The aggregation function .count() is irrelevant here. When the index is reset, 'category' becomes a column. Now, randomly sampling from this two-row grouped DataFrame such that each row is equally likely to be selected does correspond to choosing items such that each item is equally likely to be a bed or outdoor furniture item. The last line simply calculates the proportion of outdoor furniture items minus the proportion of beds in our random sample drawn according to the null model.

Difficulty: ⭐️⭐️⭐️

The average score on this problem was 59%.

Answer: <=

To answer this question, we need to know whether small values or large values of the test statistic indicate the alternative hypothesis. The alternative hypothesis is that IKEA sells more beds than outdoor furniture. Since we’re calculating the proportion of outdoor furniture minus the proportion of beds, this difference will be small (negative) if the alternative hypothesis is true. Larger (positive) values of the test statistic mean that IKEA sells more outdoor furniture than beds. A value near 0 means they sell beds and outdoor furniture equally.

The p-value is defined as the proportion of simulated test statistics that are equal to the observed value or more extreme, where extreme means in the direction of the alternative. In this case, since small values of the test statistic indicate the alternative hypothesis, the correct answer is <=.

Difficulty: ⭐️⭐️⭐️⭐️

The average score on this problem was 43%.

Answer: 0.2

Recall, the TVD is the sum of the absolute differences in proportions, divided by 2. The absolute differences in proportions for each category are as follows:

• Free Throws: |0.05 - 0.2| = 0.15
• Threes: |0.35 - 0.2| = 0.15
• Midrange: |0.35 - 0.3| = 0.05
• Layups: |0.25 - 0.3| = 0.05

Then, we have

\text{TVD} = \frac{1}{2} (0.15 + 0.15 + 0.05 + 0.05) = 0.2

Difficulty: ⭐️⭐️

The average score on this problem was 84%.

Answers: 1, np.arange(1, 50), most_sim_player = player

Let’s try and understand the code provided to us. It appears that we’re looping over the names of all other players, each time computing the TVD between Kelsey Plum’s shot distribution and that player’s shot distribution. If the TVD calculated in an iteration of the for-loop (player_tvd) is less than the previous lowest TVD (lowest_tvd_so_far), the current player (player) is now the most “similar” to Kelsey Plum, and so we store their TVD and name (in most_sim_player).

Before the for-loop, we haven’t looked at any other players, so we don’t have values to store in most_sim_player and lowest_tvd_so_far. On the first iteration of the for-loop, both of these values need to be updated to reflect Kelsey Plum’s similarity with the first player in other_players. This is because, if we’ve only looked at one player, that player is the most similar to Kelsey Plum. most_sim_player is already initialized as an empty string, and we will specify how to “update” most_sim_player in blank (c). For blank (a), we need to pick a value of lowest_tvd_so_far that we can guarantee will be updated on the first iteration of the for-loop. Recall, TVDs range from 0 to 1, with 0 meaning “most similar” and 1 meaning “most different”. This means that no matter what, the TVD between Kelsey Plum’s distribution and the first player’s distribution will be less than 1*, and so if we initialize lowest_tvd_so_far to 1 before the for-loop, we know it will be updated on the first iteration.

• It’s possible that the TVD between Kelsey Plum’s shot distribution and the first other player’s shot distribution is equal to 1, rather than being less than 1. If that were to happen, our code would still generate the correct answer, but lowest_tvd_so_far and most_sim_player wouldn’t be updated on the first iteration. Rather, they’d be updated on the first iteration where player_tvd is strictly less than 1. (We’d expect that the TVDs between all pairs of players are neither exactly 0 nor exactly 1, so this is not a practical issue.) To avoid this issue entirely, we could change if player_tvd < lowest_tvd_so_far to if player_tvd <= lowest_tvd_so_far, which would make sure that even if the first TVD is 1, both lowest_tvd_so_far and most_sim_player are updated on the first iteration.
• Note that we could have initialized lowest_tvd_so_far to a value larger than 1 as well. Suppose we initialized it to 55 (an arbitrary positive integer). On the first iteration of the for-loop, player_tvd will be less than 55, and so lowest_tvd_so_far will be updated.

Then, we need other_players to be an array containing the names of all players other than Kelsey Plum, whose name is stored at position 0 in breakdown.columns. We are told that there are 50 players total, i.e. that there are 50 columns in breakdown. We want to take the elements in breakdown.columns at positions 1, 2, 3, …, 49 (the last element), and the call to np.arange that generates this sequence of positions is np.arange(1, 50). (Remember, np.arange(a, b) does not include the second integer!)

In blank (c), as mentioned in the explanation for blank (a), we need to update the value of most_sim_player. (Note that we only arrive at this line if player_tvd is the lowest pairwise TVD we’ve seen so far.) All this requires is most_sim_player = player, since player contains the name of the player who we are looking at in the current iteration of the for-loop.

Difficulty: ⭐️⭐️⭐️

The average score on this problem was 70%.

Answer: 6

The TVD is the sum of the absolute differences in proportions, divided by 2. In the code to the left of the blank, we’ve computed the mean of the absolute differences in proportions, which is the same as the sum of the absolute differences in proportions, divided by 12 (since len(dist1) is 12). To correct the fact that we divided by 12, we multiply by 6, so that we’re only dividing by 2.

Difficulty: ⭐️⭐️⭐️⭐️⭐️

The average score on this problem was 17%.

Answer: np.ones(12) / 12

uniform_dist needs to be the same as the uniform distribution provided in the null hypothesis, \left[\frac{1}{12}, \frac{1}{12}, ..., \frac{1}{12}\right].

In code, this is an array of length 12 in which each element is equal to 1 / 12. np.ones(12) creates an array of length 12 in which each value is 1; for each value to be 1 / 12, we divide np.ones(12) by 12.

Difficulty: ⭐️⭐️⭐️

The average score on this problem was 66%.

Answer: np.random.multinomial(total_count, uniform_dist) / total_count

The idea here is to repeatedly generate an array of proportions that results from distributing total_count hours across the 12 months in a way that each month is equally likely to be chosen. Each time we generate such an array, we’ll determine its TVD from the uniform distribution; doing this repeatedly gives us an empirical distribution of the TVD under the assumption the null hypothesis is true.

Difficulty: ⭐️⭐️⭐️⭐️⭐️

The average score on this problem was 21%.

Answer: uniform_dist

As mentioned above:

Each time we generate such an array, we’ll determine its TVD from the uniform distribution; doing this repeatedly gives us an empirical distribution of the TVD under the assumption the null hypothesis is true.

Difficulty: ⭐️⭐️⭐️

The average score on this problem was 54%.

Answer: >=

The purpose of the last line of code is to compute the p-value for the hypothesis test. Recall, the p-value of a hypothesis test is the proportion of simulated test statistics that are as or more extreme than the observed test statistic, under the assumption the null hypothesis is true. In this context, “as extreme or more extreme” means the simulated TVD is greater than or equal to the observed TVD (since larger TVDs mean “more different”).

Difficulty: ⭐️⭐️

The average score on this problem was 77%.

Answer: observed_counts / total_count or observed_counts / observed_counts.sum()

Blank (f) needs to contain the observed distribution of sunshine hours (as an array of proportions) that we compare against the uniform distribution to calculate the observed TVD. This observed TVD is then compared with the distribution of simulated TVDs to calculate the p-value. The observed counts are converted to proportions by dividing by the total count so that the observed distribution is on the same scale as the simulated and expected uniform distributions, which are also in proportions.

Difficulty: ⭐️⭐️⭐️⭐️⭐️

The average score on this problem was 27%.

Answer: Any difference between these two averages is due to random chance.

If the null hypothesis is true, this means that the time recorded in app_data for each BILLY bookcase is a random number that comes from some distribution, and the time recorded in app_data for each LOMMARP bookcase is a random number that comes from the same distribution. Each assembly time is a random number, so even if the null hypothesis is true, if we take one person who assembles a BILLY bookcase and one person who assembles a LOMMARP bookcase, there is no guarantee that their assembly times will match. Their assembly times might match, or they might be different, because assembly time is random. Randomness is the only reason that their assembly times might be different, as the null hypothesis says there is no systematic difference in assembly times between the two bookcases. Specifically, it’s not the case that one typically takes longer to assemble than the other.

With those points in mind, let’s go through the answer choices.

The first answer choice is incorrect. Just because two sets of numbers are drawn from the same distribution, the numbers themselves might be different due to randomness, and the averages might also be different. Maybe just by chance, the people who assembled the BILLY bookcases and recorded their times in app_data were slower on average than the people who assembled LOMMARP bookcases. If the null hypothesis is true, this difference in average assembly time should be small, but it very likely exists to some degree.

The second answer choice is correct. If the null hypothesis is true, the only reason for the difference is random chance alone.

The third answer choice is incorrect for the same reason that the second answer choice is correct. If the null hypothesis is true, any difference must be explained by random chance.

The fourth answer choice is incorrect. If there is a difference between the averages, it should be very small and not statistically significant. In other words, if we did a hypothesis test and the null hypothesis was true, we should fail to reject the null.

Difficulty: ⭐️⭐️

The average score on this problem was 77%.

Answer: (billy_lommarp.get('minutes').sum() - billy_mean * billy.sum())/lommarp.sum()

The first line of code creates a boolean Series with a True value for every BILLY bookcase, and the second line of code creates the analogous Series for the LOMMARP bookcase. The third line queries to define a DataFrame called billy_lommarp containing all products that are BILLY or LOMMARP bookcases. In other words, this DataFrame contains a mix of BILLY and LOMMARP bookcases.

From this point, the way we would normally proceed in a permutation test would be to use np.random.permutation to shuffle one of the two relevant columns (either 'product' or 'minutes') to create a random pairing of assembly times with products. Then we would calculate the average of all assembly times that were randomly assigned to the label BILLY. Similarly, we’d calculate the average of all assembly times that were randomly assigned to the label LOMMARP. Then we’d subtract these averages to get one simulated value of the test statistic. To run the permutation test, we’d have to repeat this process many times.

In this problem, we need to generate a simulated value of the test statistic, without randomly shuffling one of the columns. The code starts us off by defining a variable called billy_mean that comes from using np.random.choice. There’s a lot going on here, so let’s break it down. Remember that the first argument to np.random.choice is a sequence of values to choose from, and the second is the number of random choices to make. And we set replace=False, so that no element that has already been chosen can be chosen again. Here, we’re making our random choices from the 'minutes' column of billy_lommarp. The number of choices to make from this collection of values is billy.sum(), which is the sum of all values in the billy Series defined in the first line of code. The billy Series contains True/False values, but in Python, True counts as 1 and False counts as 0, so billy.sum() evaluates to the number of True entries in billy, which is the number of BILLY bookcases recorded in app_data. It helps to think of the random process like this:

1. Collect all the assembly times of any BILLY or LOMMARP bookcase in a large bag.
2. Pull out a random assembly time from this bag.
3. Repeat step 2, drawing as many times as there are BILLY bookcases, without replacement.

If we think of the random times we draw as being labeled BILLY, then the remaining assembly times still leftover in the bag represent the assembly times randomly labeled LOMMARP. In other words, this is a random association of assembly times to labels (BILLY or LOMMARP), which is the same thing we usually accomplish by shuffling in a permutation test.

From here, we can proceed the same way as usual. First, we need to calculate the average of all assembly times that were randomly assigned to the label BILLY. This is done for us and stored in billy_mean. We also need to calculate the average of all assembly times that were randomly assigned the label LOMMARP. We’ll call that lommarp_mean. Thinking of picking times out of a large bag, this is the average of all the assembly times left in the bag. The problem is there is no easy way to access the assembly times that were not picked. We can take advantage of the fact that we can easily calculate the total assembly time of all BILLY and LOMMARP bookcases together with billy_lommarp.get('minutes').sum(). Then if we subtract the total assembly time of all bookcases randomly labeled BILLY, we’ll be left with the total assembly time of all bookcases randomly labeled LOMMARP. That is, billy_lommarp.get('minutes').sum() - billy_mean * billy.sum() represents the total assembly time of all bookcases randomly labeled LOMMARP. The count of the number of LOMMARP bookcases is given by lommarp.sum() so the average is (billy_lommarp.get('minutes').sum() - billy_mean * billy.sum())/lommarp.sum().

A common wrong answer for this question was the second answer choice, np.random.choice(billy_lommarp.get('minutes'), lommarp.sum(), replace=False).mean(). This mimics the structure of how billy_mean was defined so it’s a natural guess. However, this corresponds to the following random process, which doesn’t associate each assembly with a unique label (BILLY or LOMMARP):

1. Collect all the assembly times of any BILLY or LOMMARP bookcase in a large bag.
2. Pull out a random assembly time from this bag.
3. Repeat, drawing as many times as there are BILLY bookcases, without replacement.
4. Collect all the assembly times of any BILLY or LOMMARP bookcase in a large bag.
5. Pull out a random assembly time from this bag.
6. Repeat step 5, drawing as many times as there are LOMMARP bookcases, without replacement.

We could easily get the same assembly time once for BILLY and once for LOMMARP, while other assembly times could get picked for neither. This process doesn’t split the data into two random groups as desired.

Difficulty: ⭐️⭐️⭐️⭐️⭐️

The average score on this problem was 12%.

Answer: Option 4, Option 6

For blank (a), we want to choose a test statistic that helps us distinguish between the null and alternative hypotheses. The alternative hypothesis says that denied and approved should be different, but it doesn’t say which should be larger. Options 1 through 3 therefore won’t work, because high values and low values of these statistics both point to the alternative hypothesis, and moderate values point to the null hypothesis. Options 4 through 6 all work because large values point to the alternative hypothesis, and small values close to 0 suggest that the null hypothesis should be true.

For blank (b), we want to calculate the p-value in such a way that it represents the proportion of trials for which the simulated test statistic was equal to the observed statistic or further in the direction of the alternative. For all of Options 4 through 6, large values of the test statistic indicate the alternative, so we need to calculate the p-value with a >= sign, as in Options 4 and 6.

While Option 3 filled in blank (a) correctly, it did not fill in blank (b) correctly. Options 4 and 6 fill in both blanks correctly.

Difficulty: ⭐️⭐️

The average score on this problem was 78%.

Answer: Option 1

As in the previous part, we need to fill blank (a) with a test statistic such that large values point towards one of the hypotheses and small values point towards the other. Here, the alterntive hypothesis suggests that approved should be less than denied, so we can’t use Options 4 through 6 because these can only detect whether approved and denied are not different, not which is larger. Any of Options 1 through 3 should work, however. For Options 1 and 2, large values point towards the alternative, and for Option 3, small values point towards the alternative. This means we need to calculate the p-value in blank (b) with a >= symbol for the test statistic from Options 1 and 2, and a <= symbol for the test statistic from Option 3. Only Options 1 fills in blank (b) correctly based on the test statistic used in blank (a).

Difficulty: ⭐️⭐️

The average score on this problem was 83%.

Answer: np.abs(stats) >= np.abs(test_stat(apps))

First, we need to understand how Option 6 works. Option 6 produces large values of the test statistic when approved is very different from denied, then calculates the p-value as the proportion of trials for which the simulated test statistic was larger than the observed statistic. In other words, Option 6 calculates the proportion of trials in which approved and denied are more different in a pair of random samples than they are in the original samples.

For Option 7, the test statistic for a pair of random samples may come out very large or very small when approved is very different from denied. Similarly, the observed statistic may come out very large or very small when approved and denied are very different in the original samples. We want to find the proportion of trials in which approved and denied are more different in a pair of random samples than they are in the original samples, which means we want the proportion of trials in which the absolute value of approved - denied in a pair of random samples is larger than the absolute value of approved - denied in the original samples.

Difficulty: ⭐️⭐️⭐️

The average score on this problem was 56%.

Answer: The blanks should be filled in as follows:

• 1. df.sample(df.shape[0])
• 2. df.take(np.arange(denied))
• 3. df.take(np.arange(denied, df.shape[0]))

For blank (a), we are told to return a DataFrame with the same rows but in a different order. We can use the .sample method for this question. We want each row of the input DataFrame df to appear once, so we should sample without replacement, and we should have has many rows in the output as in df, so our sample should be of size df.shape[0]. Since sampling without replacement is the default behavior of .sample, it is optional to specify replace=False.

Difficulty: ⭐️⭐️⭐️

The average score on this problem was 59%.

For blank (b), we need to implement the strategy outlined, where after we shuffle the DataFrame, we use the values at the top of the DataFrame as our new “denied sample. In a permutation test, the two random groups we create should have the same sizes as the two original groups we are given. In this case, the size of the”denied” group in our original data is stored in the variable denied. So we need the rows in positions 0, 1, 2, …, denied - 1, which we can get using df.take(np.arange(denied)).

Difficulty: ⭐️⭐️⭐️⭐️

The average score on this problem was 39%.

For blank (c), we need to get all remaining applicants, who form the new “approved” sample. We can .take the rows corresponding to the ones we didn’t put into the “denied” group. That is, the first applicant who will be put into this group is at position denied, and we’ll take all applicants from there onwards. We should therefore fill in blank (c) with df.take(np.arange(denied, df.shape[0])).

For example, if apps had only 10 rows, 7 of them corresponding to denied applications, we would shuffle the rows of apps, then take rows 0, 1, 2, 3, 4, 5, 6 as our new “denied” sample and rows 7, 8, 9 as our new “approved” sample.

Difficulty: ⭐️⭐️⭐️⭐️

The average score on this problem was 38%.

Answer: This is a permutation test, and our test statistic is the difference in the mean Adelie mass and mean Chinstrap mass (Option 4)

Recall, a permutation test helps us decide whether two random samples come from the same distribution. This test matches our goal of testing whether the masses of Adelie penguins and Chinstrap penguins are drawn from the same population distribution. The code above are also doing steps of a permutation test. In part (a), it shuffles 'species' and stores the shuffled series to shuffled. In part (b), it assign the shuffled series of values to 'species' column. Then, it uses grouped = with_shuffled.groupby('species').mean() to calculate the mean of each species. In part (c), it computes the difference between mean mass of the two species by first getting the 'mass' column and then accessing mean mass of each group (Adelie and Chinstrap) with positional index 0 and 1.

Difficulty: ⭐️

The average score on this problem was 98%.

Answer: This would not run a valid hypothesis test, as all values in the stats array would be exactly the same (Option 2)

Recall, DataFrame.sample(n, replace = False) (or DataFrame.sample(n) since replace = False is by default) returns a DataFrame by randomly sampling n rows from the DataFrame, without replacement. Since our n is adelie_chinstrap.shape[0], and we are sampling without replacement, we will get the exactly same Dataframe (though the order of rows may be different but the stats array would be exactly the same).

Difficulty: ⭐️⭐️

The average score on this problem was 87%.

Answer: This would not run a valid hypothesis test, even though there would be several different values in the stats array (Option 3)

Recall, DataFrame.sample(n, replace = True) returns a new DataFrame by randomly sampling n rows from the DataFrame, with replacement. Since we are sampling with replacement, we will have a DataFrame which produces a stats array with some different values. However, recall, the key idea behind a permutation test is to shuffle the group labels. So, the above code does not meet this key requirement since we only want to shuffle the "species" column without changing the size of the two species. However, the code may change the size of the two species.

Difficulty: ⭐️⭐️⭐️

The average score on this problem was 66%.

Answer: This would still run a valid hypothesis test (Option 1)

Our goal for the permutation test is to randomly assign birth weights to groups, without changing group sizes. The above code shuffles 'species' and 'mass' columns and assigns them back to the DataFrame. This fulfills our goal.

Difficulty: ⭐️⭐️

The average score on this problem was 81%.

Answer: \frac{1}{3}

Recall, the p-value is the chance, under the null hypothesis, that the test statistic is equal to the value that was observed in the data or is even further in the direction of the alternative. Thus, we compute the proportion of the test statistic that is equal or less than the observed statistic. (It is less than because less than corresponds to the alternative hypothesis “Chinstrap penguins weigh more on average than Adelie penguins”. Recall, when computing the statistic, we use Adelie’s mean mass minus Chinstrap’s mean mass. If Chinstrap’s mean mass is larger, the statistic will be negative, the direction of less than the observed statistic).

Thus, we look at the proportion of area less than or on the red line (which represents observed statistic), it is around \frac{1}{3}.

Difficulty: ⭐️⭐️

The average score on this problem was 80%.

Anwser: Hypothesis Testing

This is a question of whether a certain set of incomes (corresponding to applicants with 2 or fewer dependents) are drawn randomly from a certain population (incomes of all applicants). We need to use hypothesis testing to determine whether this model for how samples are drawn from a population seems plausible.

Difficulty: ⭐️⭐️⭐️⭐️

The average score on this problem was 47%.

Anwser: Bootstrapping

The question is looking for an estimate a specific parameter (the median income of applicants with 2 or fewer dependents), so we know boostrapping is the best tool.

Difficulty: ⭐️⭐️

The average score on this problem was 88%.

Anwser: Hypothesis Testing

The question asks about the validity of a model in which applications are approved randomly such that each application has a 50% chance of being approved. To determine whether this model is plausible, we should use a standard hypothesis test to simulate this random process many times and see if the data generated according to this model is consistent with our observed data.

Difficulty: ⭐️⭐️⭐️

The average score on this problem was 74%.

Anwser: Permutation Testing

Recall, a permutation test helps us decide whether two random samples come from the same distribution. This question is about whether two random samples for different groups of applicants have the same distribution of incomes or whether they don’t because one group’s median incomes is less than the other.

Difficulty: ⭐️⭐️⭐️

The average score on this problem was 57%.

Anwser: Bootstrapping

The question at hand is looking for a specific parameter value (the difference in median incomes for two different subsets of the applicants). Since this is a question of estimating an unknown parameter, bootstrapping is the best tool.

Difficulty: ⭐️⭐️⭐️

The average score on this problem was 63%.