QUANTITATIVE METHODS – I ASSIGNMENT

DEC 2025

Question No. 1
Answer:
1. Introduction
In business analytics and quantitative methods, organizations often rely on sample data to make
inferences about a large population. It is usually impractical to collect data from every customer, so
analysts use statistical estimation techniques to predict population characteristics.
In this scenario, a telecommunications company has launched a pilot survey for its new internet
service. Out of 250 customers surveyed, 162 expressed interest in subscribing. The marketing
analyst must estimate, with 90% confidence, the true proportion of all customers likely to be
interested in the service.
The objective here is to apply the interval estimation method for proportions to calculate a
confidence interval that represents the likely range of customer interest in the population.
2. Concept of Confidence Interval for Proportion
A confidence interval (CI) provides a range of values within which the true population parameter
(in this case, population proportion) is expected to lie, with a specified level of confidence.
When the sample size is large, the sampling distribution of the sample proportion (𝑝̂ ) follows an
approximately normal distribution, as stated by the Central Limit Theorem.
The general formula for the confidence interval of a proportion is:

𝑝̂ (1 − 𝑝̂ )
𝐶𝐼 = 𝑝̂ ± 𝑍𝛼/2 × √
𝑛

where:
• 𝑝̂ = sample proportion = 𝑥/𝑛
• 𝑥= number of “successes” or customers showing interest
• 𝑛= sample size
• 𝑍𝛼/2= Z-value from the standard normal distribution corresponding to the desired
confidence level
• 𝛼= 1 – confidence level
This formula allows estimation of the range of true customer interest in the population.
3. Step-by-Step Application
Step 1: Identify Given Data

QUANTITATIVE METHODS – I 1
Parameter Symbol Value

Sample size n 250

Number of interested customers x 162

Confidence level — 90%

Compute the sample proportion (𝑝̂ ):

𝑥 162
𝑝̂ = = = 0.648
𝑛 250

Hence, the sample proportion = 0.648 (64.8%).

Step 2: Determine Z-value
For a 90% confidence level:
𝛼 = 1 − 0.90 = 0.10 ⇒ 𝛼/2 = 0.05

From the standard normal distribution table,

𝑍0.05 = 1.645

Step 3: Compute Standard Error (SE)

The Standard Error (SE) measures the variability of the sample proportion:

𝑝̂ (1 − 𝑝̂ )
𝑆𝐸 = √
𝑛

Substitute the values:

0.648 × (1 − 0.648) 0.648 × 0.352

𝑆𝐸 = √ =√
250 250

0.228096
𝑆𝐸 = √ = √0.000912384 = 0.0302
250

Thus, SE = 0.0302.
Step 4: Calculate Margin of Error (ME)
𝑀𝐸 = 𝑍𝛼/2 × 𝑆𝐸 = 1.645 × 0.0302 = 0.0497

Hence, the Margin of Error = 0.0497 (4.97%).

Step 5: Construct the Confidence Interval

QUANTITATIVE METHODS – I 2
 𝐶𝐼 = 𝑝̂ ± 𝑀𝐸 = 0.648 ± 0.0497

• Lower limit = 0.648 – 0.0497 = 0.598

• Upper limit = 0.648 + 0.0497 = 0.698
Therefore,

90% Confidence Interval = (0.598, 0 .698)

In percentage terms: 59.8% to 69.8%

4. Interpretation of Results
The interpretation of this interval is as follows:
• The marketing analyst can be 90% confident that the true proportion of all customers who
are interested in subscribing to the new internet service lies between 59.8% and 69.8%.
• In other words, if the company conducted many similar random surveys, 90% of the
resulting intervals would contain the true population proportion.
• The sample estimate of 64.8% is the best single-point estimate, while the interval
provides the range of uncertainty around it.
5. Checking Conditions for Validity
For the normal approximation to hold, the following conditions must be met:
𝑛𝑝̂ ≥ 5and𝑛(1 − 𝑝̂ ) ≥ 5

Substitute values:
𝑛𝑝̂ = 250 × 0.648 = 162
𝑛(1 − 𝑝̂ ) = 250 × 0.352 = 88

Both are greater than 5, hence conditions satisfied.

Therefore, the normal approximation is valid and appropriate in this scenario.
6. Managerial Interpretation and Business Implications
This confidence interval is a crucial decision-support tool for the company’s marketing and product
strategy. It helps in quantifying uncertainty and making data-driven decisions.
a) Assessing Market Potential
Even at the lower bound (59.8%), more than half of the customers show interest. This indicates a
strong potential market for the new service, suggesting a favorable response.
b) Measuring Reliability
The margin of error (±4.97%) indicates a moderate degree of precision. For critical business
launches, management might aim to reduce this margin by increasing the sample size or
conducting additional surveys.

QUANTITATIVE METHODS – I 3
c) Marketing and Pricing Strategy
If 60–70% of customers are expected to be interested, the company can plan pricing models,
marketing campaigns, and network capacity to meet potential demand effectively.
d) Decision-Making Insight
Since the 65% interest level (the point estimate) lies within the confidence interval, management
can be reasonably confident that the majority of customers are open to the new service,
supporting further investment.
7. Limitations of Confidence Interval Approach
While confidence intervals provide valuable insights, the analyst must be aware of the limitations:
1. Random Sampling Requirement: The sample must be randomly selected; any bias in
selection affects the validity of the interval.
2. Sample Size Impact: Smaller samples lead to wider intervals and less precise estimates.
3. Confidence Level Trade-Off: Higher confidence (e.g., 95%) widens the interval, reducing
precision but increasing reliability.
4. Assumption of Normality: The method assumes normal approximation, which might not
hold for very small samples or extreme proportions.
5. Sampling Error: Confidence intervals do not account for non-sampling errors such as
inaccurate responses or survey bias.
8. Recommendations for Further Analysis
To strengthen future research and ensure better strategic decisions, the following actions are
recommended:
• Increase Sample Size: A larger sample reduces the margin of error, giving a more precise
estimate.
• Segmented Surveys: Analyze interest levels across customer demographics (age,
location, income) for targeted marketing.
• Track Over Time: Conduct follow-up surveys after marketing campaigns to monitor shifts in
interest levels.
• Cost–Benefit Analysis: Use the confidence interval range to forecast potential
subscription revenues and assess profitability before rollout.
9. Conclusion
To conclude, the marketing analyst successfully applied the interval estimation formula for
proportions to estimate the proportion of customers interested in the new internet service. Based
on the survey results:
\boxed{\text{90% Confidence Interval} = 59.8\% \text{ to } 69.8\%}
This means the company can be 90% confident that the true customer interest level falls within
this range.

QUANTITATIVE METHODS – I 4
The result suggests that a majority of customers are likely to subscribe, indicating strong
market potential for the new service. However, to make final strategic decisions, management
should consider expanding the sample size and performing segment-wise analyses.
Through this example, it is evident how quantitative methods and statistical inference provide
actionable business insights and reduce uncertainty in managerial decision-making.
Final Answer Summary

Element Value

Sample size (n) 250

Interested customers (x) 162

Sample proportion (𝑝̂ ) 0.648

Confidence Level 90%

Z-value 1.645

Standard Error 0.0302

Margin of Error 0.0497

90% Confidence Interval (0.598, 0.698) or (59.8%, 69.8%)

Question No. 2 A)
Answer:
Applying Bayes’ theorem in a financial advisory firm to update probabilities after receiving new
information on client satisfaction is an appropriate and statistically valid approach. Bayes’ theorem
helps in revising prior probabilities (existing beliefs) in light of new evidence (such as a client’s
feedback) to obtain updated or posterior probabilities. In this case, the firm can use it to estimate
the probability that a highly satisfied client was served by each advisor, based on prior client
distribution and historical satisfaction rates.
Mathematically, Bayes’ theorem states:
𝑃(𝐻 ∣ 𝐴𝑖 ) × 𝑃(𝐴𝑖 )
𝑃(𝐴𝑖 ∣ 𝐻) =
∑𝑃(𝐻 ∣ 𝐴𝑗 ) × 𝑃(𝐴𝑗 )

where 𝑃(𝐴𝑖 )is the prior probability that a client belongs to advisor 𝐴𝑖 , 𝑃(𝐻 ∣ 𝐴𝑖 )is the probability of
high satisfaction given that advisor 𝐴𝑖 served the client, and 𝑃(𝐴𝑖 ∣ 𝐻)is the posterior probability
after observing the client’s high satisfaction.
Appropriateness of the approach:
Using Bayes’ theorem is appropriate because it allows the firm to update beliefs logically as new
satisfaction data become available. It integrates both the existing knowledge (such as number of
clients per advisor) and new feedback evidence, thereby improving decision-making accuracy. For
example, even if one advisor handles fewer clients, if their clients consistently report higher
satisfaction, Bayes’ theorem will adjust the posterior probability upward, reflecting better
QUANTITATIVE METHODS – I 5
performance. This helps management identify top-performing advisors and allocate future clients
or training resources more effectively.
However, the usefulness of Bayes’ results depends on several factors:
1. Validity of Prior Probabilities: The priors should accurately reflect the current client base
of each advisor. Outdated or incorrect client data will distort posterior results.
2. Accuracy of Likelihoods: The conditional probabilities 𝑃(𝐻 ∣ 𝐴𝑖 )—the satisfaction rates for
each advisor—must be estimated from reliable, unbiased data. If the satisfaction survey is
inconsistent across advisors, or if response rates vary, the results will be misleading.
3. Sample Size and Representativeness: The firm should ensure that satisfaction data are
collected from a large and representative sample. Small or biased samples can produce
unstable posterior probabilities.
4. Control of Confounding Variables: Client satisfaction may also depend on factors beyond
the advisor’s control, such as investment outcomes, client expectations, or market
conditions. The firm should adjust for these factors before drawing conclusions about
advisor performance.
5. Dynamic Nature of Data: Since client satisfaction can change over time, the firm should
periodically update the priors and likelihoods using recent data, ensuring the results remain
current and reflective of real performance.
6. Ethical and Practical Considerations: While Bayes’ theorem provides a scientific basis
for updating probabilities, the firm must be cautious when using these results for
performance evaluation. Overreliance on statistical probabilities without considering
qualitative aspects like client rapport, service quality, or teamwork could lead to unfair
assessments.
Conclusion:
In summary, applying Bayes’ theorem is an appropriate and powerful method for updating the
firm’s understanding of advisor performance based on new client satisfaction data. It supports
evidence-based decision-making and helps management allocate resources more effectively.
However, the firm must ensure accurate data, unbiased collection methods, and consideration of
other qualitative factors. When used carefully alongside managerial judgment, Bayes’ theorem can
provide meaningful and actionable insights into advisor performance and client satisfaction trends.

Question No. 2 B)
Answer:
In a large financial institution, standardizing risk analysis procedures is essential for maintaining
consistency, accuracy, and reliability across departments. When dealing with normal distribution
calculations, two common approaches are available: using Excel’s [Link] and [Link]
functions, or relying on the traditional z-table method. Each approach has its own implications in
terms of computational efficiency, accuracy, and interpretability, and the institution must carefully
evaluate these factors before setting a standard.
The traditional z-table method has long been used in statistics to determine probabilities
associated with the standard normal distribution. It helps analysts manually convert raw scores

QUANTITATIVE METHODS – I 6
into z-scores and find the corresponding probability values. This method provides a clear
conceptual understanding of how probabilities relate to standard deviations and the normal curve.
It is particularly useful for educational and interpretive purposes, as it strengthens the analyst’s
foundational understanding of statistical principles. However, it can be time-consuming and prone
to human error, especially when analysts need to perform multiple calculations or work with non-
standard normal distributions (i.e., with different means and standard deviations).
On the other hand, Excel’s [Link] and [Link] functions automate these calculations
with high precision. [Link] can quickly compute the probability associated with a given value
in a normal distribution, while [Link] can determine the value corresponding to a given
probability. These functions significantly improve computational efficiency and accuracy, reducing
the likelihood of manual errors. They also allow for easy scalability when dealing with large
datasets or repeated risk simulations—an essential aspect in modern financial analysis and risk
modeling.
However, the main drawback of relying solely on Excel functions is that some analysts may treat
these tools as “black boxes,” without fully understanding the underlying statistical logic. This can
lead to reduced interpretability and potential misapplication of results, particularly among new
analysts who may depend entirely on automated outputs. Hence, while Excel offers speed and
precision, it may compromise on conceptual clarity if not paired with proper training.
When weighing the trade-offs, the institution should prioritize accuracy and efficiency, as financial
risk assessments directly influence decision-making, regulatory compliance, and portfolio
management. Given that modern financial environments are data-driven, adopting Excel’s built-in
statistical functions as the standard method would be more appropriate. However, to maintain
interpretability and statistical literacy, the institution should integrate training programs that explain
the theory behind the normal distribution and z-scores. This ensures that analysts not only use
Excel efficiently but also understand what the functions represent.
In conclusion, the institution should standardize the use of Excel’s [Link] and [Link]
functions for all probability and risk calculations to ensure consistency, precision, and scalability. At
the same time, it should emphasize ongoing conceptual training to preserve interpretability and
sound analytical judgment. This balanced approach aligns with both operational efficiency and the
institution’s commitment to analytical excellence.

QUANTITATIVE METHODS – I 7