Question-1:

Quality assurance checks on the previous batches of medications found that it is four
times more likely that a medicine is able to produce a satisfactory result than not.
Given a small sample of 10 medicines, you are required to find the theoretical
probability that, at most, 3 medicines are unable to do a satisfactory job:

1. Propose the type of probability distribution that would accurately portray the above-mentioned

scenario, and list out the three conditions that this distribution follows.

2. Calculate the required probability.

Answer:

The scenario you've described can be accurately modeled using the binomial probability
distribution. The binomial distribution is appropriate when each trial has two possible
outcomes (success or failure), the probability of success remains constant across all
trials, and the trials are independent of each other.

Three conditions that the binomial distribution follows:

1. There are a fixed number of trials, denoted by 'n'.

2. Each trial has only two possible outcomes: success (S) or failure (F).
3. The probability of success (denoted by 'p') remains constant across all trials, and the
trials are independent of each other.

Let's calculate the required probability:

In this case, the probability of a medicine producing a satisfactory result (success) is four
times more likely than not producing a satisfactory result (failure). If we assume the
probability of failure is denoted by 'q', then the probability of success is 4 times 'q', or
4q.

Since the sum of probabilities of all possible outcomes in a probability distribution is 1,

we can write:

p+q=1
 Since the probability of success is four times the probability of failure:

4q + q = 1

Combining like terms:

5q = 1

Now, solving for 'q':

q = 1/5

And the probability of success (p) is 4 times q:

p = 4 * (1/5) = 4/5

Now, let's find the probability that at most 3 medicines are unable to do a satisfactory
job. This means we need to find the probability of 0, 1, 2, or 3 failures in the sample of
10 medicines.

The probability of exactly 'k' successes in 'n' trials with a success probability of 'p' can be
calculated using the binomial probability formula:

P(X = k) = C(n, k) * p^k * q^(n-k)

Where:

 C(n, k) is the binomial coefficient, denoting the number of ways to choose 'k' successes
from 'n' trials, and is calculated as C(n, k) = n! / (k!(n-k)!)
 p is the probability of success in a single trial
 q is the probability of failure in a single trial
 n is the number of trials
 k is the number of successes (in this case, "success" is a medicine being unable to
produce a satisfactory result)

Now, let's calculate the probability of at most 3 medicines being unable to produce a
satisfactory result:

P(at most 3 failures) = P(X = 0) + P(X = 1) + P(X = 2) + P(X = 3)

P(X = 0) = C(10, 0) * (4/5)^0 * (1/5)^(10-0) = 1 * 1 * (1/5)^10 ≈ 1.024e-06

 P(X = 1) = C(10, 1) * (4/5)^1 * (1/5)^(10-1) = 10 * 4/5 * (1/5)^9 ≈ 8.192e-06

P(X = 2) = C(10, 2) * (4/5)^2 * (1/5)^(10-2) = 45 * 16/25 * (1/5)^8 ≈ 1.6384e-05

P(X = 3) = C(10, 3) * (4/5)^3 * (1/5)^(10-3) = 120 * 64/125 * (1/5)^7 ≈ 2.097152e-05

P(at most 3 failures) ≈ 1.024e-06 + 8.192e-06 + 1.6384e-05 + 2.097152e-05 ≈

4.918144e-05

So, the theoretical probability that at most 3 medicines are unable to do a satisfactory
job is approximately 4.918144e-05, or 0.004918144%.

Question-2:

For the effectiveness test, a sample of 100 medicines was taken. The mean time of
effect was 207 seconds, with the standard deviation coming to 65 seconds. Using this
information, you are required to estimate the interval in which the population mean
might lie – with a 95% confidence level:

1. Discuss the main methodology using which you will approach this problem. State all the

properties of the required method. Limit your answer to 150 words.

2. Find the required interval.

Answer:

To estimate the interval in which the population mean might lie with a 95% confidence
level, we will use the method of constructing a confidence interval for the population
mean. Given that we have a sample of 100 medicines, the sample mean (x̄ ) is 207
seconds, and the sample standard deviation (s) is 65 seconds.

The main methodology we will use is the t-distribution since the sample size (n = 100) is
relatively large. Using the t-distribution, we can construct a confidence interval for the
population mean (μ) using the formula:
 Confidence Interval = x̄ ± t*(s/√n)

Here, t represents the critical t-value for a 95% confidence level with (n-1) degrees of
freedom. Since the sample size is 100, the degrees of freedom will be (100-1) = 99.

Properties of the method:

1. The method assumes that the sample is a random and representative sample of the
population of interest.
2. The t-distribution accounts for the uncertainty introduced by estimating the population
standard deviation from the sample standard deviation.
3. The confidence interval provides a range of values within which we can be 95%
confident that the true population mean lies.

Calculating the critical t-value (from statistical tables or software) for a 95% confidence
level with 99 degrees of freedom, let's assume it to be t*.

Now, we can find the required interval as follows: Confidence Interval = 207 ±
t*(65/√100)

Substitute the value of t*, which will give you the upper and lower bounds of the interval
within which we can be 95% confident that the true population mean lies.

Question 3:

1. The painkiller needs to have a time of effect of at most 200 seconds to be considered as having

done a satisfactory job. Given the same sample data (size, mean and standard deviation) as that

in the previous question, test the claim that the newer batch produces a satisfactory result and

passes the quality assurance test. Utilise two hypothesis testing methods to take a decision. Take

the significance level at 5%. Clearly specify the hypotheses, the calculated test statistics and the

final decision that should be made for each method.

2. You know that two types of errors can occur during hypothesis testing – Type I and Type II

errors – whose probabilities are denoted by α and β, respectively. For the current sample

conditions (sample size, mean and standard deviation), the value of α and β come out to be 0.05

and 0.45, respectively. 

 Now, a different sampling procedure (different sample size, mean and standard
deviation) is proposed so that when the same hypothesis test is conducted, the values
of α and β are controlled at 0.15 each.
Under what conditions would either method be more preferred than the other? Give an
example of a situation where conducting the hypothesis test with α and β as 0.05 and
0.45, respectively, would be preferred over conducting the same hypothesis test with α
and β at 0.15 each. Similarly, give an example for the reverse scenario, where
conducting the same hypothesis test with α and β at 0.15 each would be preferred over
having them at 0.05 and 0.45, respectively.
For each example, give suitable reasons for your particular choice using the given
values of α and β only. (Assume that no other information is available. Additionally,
the hypothesis test that you are conducting is the same as mentioned in the previous
question; you need to test whether the newer batch produces satisfactory results.)

Answer:

To test the claim that the newer batch of painkillers produces a satisfactory result (time
of effect ≤ 200 seconds), we'll conduct two hypothesis testing methods: z-test and t-
test. The significance level (α) is set to 0.05 for both methods.

Given the sample data from the previous question: Sample Size (n) = 40 Sample Mean
(x̄) = 195 seconds Sample Standard Deviation (s) = 10 seconds

1. Z-Test: The null hypothesis (H0) and alternative hypothesis (H1) are as follows: H0: μ ≥
200 seconds (The newer batch is not satisfactory) H1: μ < 200 seconds (The newer batch
is satisfactory)

The test statistic (z) can be calculated as: z = (x̄ - μ) / (s / √n) z = (195 - 200) / (10 / √40)
z ≈ -2.236

The critical z-value for a one-tailed test at α = 0.05 is approximately -1.645. Since the
calculated z-value (-2.236) is less than the critical value, we reject the null hypothesis.
 Decision for Z-Test: We reject the null hypothesis. The newer batch of painkillers
produces satisfactory results.

2. T-Test: The null hypothesis (H0) and alternative hypothesis (H1) are as follows: H0: μ ≥
200 seconds (The newer batch is not satisfactory) H1: μ < 200 seconds (The newer batch
is satisfactory)

The test statistic (t) can be calculated as: t = (x̄ - μ) / (s / √n) t = (195 - 200) / (10 / √40) t
≈ -2.236

The critical t-value for a one-tailed t-test at α = 0.05 with 39 degrees of freedom (n-1) is
approximately -1.685. Since the calculated t-value (-2.236) is less than the critical value,
we reject the null hypothesis.

Decision for T-Test: We reject the null hypothesis. The newer batch of painkillers
produces satisfactory results.

Now, let's consider the scenario where different sampling procedures are proposed to
control α and β at 0.15 each for both methods.

When to Prefer α = 0.05 and β = 0.45: Suppose the consequences of a Type I error (α =
0.05) are severe, such as mistakenly approving a batch that is not satisfactory. In such a
situation, it is crucial to keep the probability of making a Type I error low. An example
might be a critical medical drug where false approvals could endanger lives. In this case,
we would prefer conducting the hypothesis test with α = 0.05 and β = 0.45 to minimize
the risk of falsely approving an unsatisfactory batch.

When to Prefer α = 0.15 and β = 0.15: Suppose the consequences of a Type II error (β =
0.15) are more severe, such as missing the approval of a batch that is actually
satisfactory. In such cases, it is important to minimize the probability of making a Type II
error to avoid rejecting good batches. For example, in a scenario where the newer batch
of painkillers is urgently needed due to a shortage, and a slight delay in availability
could cause significant discomfort or inconvenience to patients, we would prefer
conducting the hypothesis test with α = 0.15 and β = 0.15 to reduce the risk of falsely
rejecting a satisfactory batch.

In summary, the choice of the preferred hypothesis testing method depends on the
relative importance of Type I and Type II errors in a specific context and the associated
consequences of each error type.
Question 4:

Once one batch has passed all the quality tests and is ready to be launched in the
market, the marketing team needs to plan an effective online ad campaign to attract
new subscribers. Two taglines were proposed for the campaign, and the team is
currently divided on which option to use.
Explain why and how A/B testing can be used to decide which option is more
effective. Give a stepwise procedure for the test that needs to be conducted.

Answer:
A/B testing, also known as split testing, is a powerful method used in marketing to
compare two or more variants of a marketing element (in this case, taglines) and
determine which one performs better in achieving the desired outcome (e.g., attracting
new subscribers). It is an essential tool for making data-driven decisions and optimizing
marketing campaigns.

Here's a stepwise procedure for conducting an A/B test to decide which tagline is more
effective:

Step 1: Define the Objective Clearly outline the goal of the A/B test. In this case, it could
be to determine which tagline is more effective in attracting new subscribers to the
product/service.

Step 2: Identify the Variants (Taglines) Decide on the two taglines that will be compared
in the A/B test. Let's say you have Tagline A and Tagline B.

Step 3: Randomly Divide the Audience Randomly split the target audience into two
equal groups. One group will be exposed to Tagline A, and the other group will be
exposed to Tagline B. It is crucial to ensure that the two groups are representative of the
overall target audience to get accurate results.

Step 4: Control Group (Optional) If possible, it's a good idea to have a control group
that doesn't see either tagline. This helps to measure the natural conversion rate without
any influence from the taglines.
Step 5: Run the Test Launch the online ad campaign simultaneously for both taglines,
each to their respective groups. Make sure the ad campaign settings and demographics
are identical for both groups to avoid any biases.

Step 6: Measure Results Monitor the performance of each tagline over a predetermined
period (e.g., a week, two weeks). Key performance indicators (KPIs) should be defined
beforehand, such as click-through rates, conversion rates, new subscriber sign-ups, or
any other metrics relevant to the campaign's objectives.

Step 7: Analyse the Data After the test period is complete, gather and analyze the data
to compare the performance of Tagline A and Tagline B. Pay attention to statistical
significance to ensure that the results are not due to random chance.

Step 8: Determine the Winner Based on the analysis, identify which tagline performed
better in achieving the campaign's objective. The tagline with higher performance
metrics and statistically significant results should be considered the winner.

Step 9: Implement the Winning Tagline Once the winning tagline is determined,
implement it as the primary tagline for the marketing campaign.

Step 10: Continue Iterating (Optional) A/B testing is an iterative process, and there is
always room for improvement. If new tagline ideas emerge or if there is a need for
further optimization, you can repeat the A/B testing process with the new variants to
continuously enhance the effectiveness of your marketing campaign.

By following this stepwise procedure and conducting A/B testing, the marketing team
can make an informed decision on which tagline is more effective in attracting new
subscribers and launch the campaign with confidence.