Solution to Management Science Series #204: Data Analytics and Forensics: Coroner’s Inquest

Data Analytics and Forensics: Coroner’s Inquest

Given the family’s demographics, the probability that the family’s first child dies of SIDS is 1 out of 1300.

Considering all other factors such as the mental state of the family and the neighborhood the family lives in, the probability that the first child is murdered is 1 out of 21,700.

Unfortunately, the family’s second child also died after 18 months had passed since first child’s death. The probability that second child dies of SIDS given that first child already died of SIDS is 1 out of 130. Furthermore, the probability that second child is murdered given that the first child was murdered is 176 out of 21,700.

A coroner assigned to this case is investigating how this could happen. The first thing she wants to look at is the probability that both first child and the second child die of SIDS given the subsequent deaths of two children. Help the coroner find out this probability.

Solution to Management Science Series #203: Real-world business problems: Improving the Customer Service at Hanamichi Sushi Restaurant

You opened the first sushi restaurant in Konoha, Hanamichi Robot Sushi, which is fully operated by robots.

The time to serve a customer is normally distributed with a mean of 8 minutes and a standard deviation of 2 minutes.

1.Find the probability that it takes 8 minutes or less to serve the next customer.

2.Find the probability that it takes 10 minutes or more to serve the next customer.

3.Find the probability that it takes 6 minutes or less to serve the next customer.

4.Do your answers to 2 and 3 resemble or differ from each other? Compare your answers.

As the owner of Hanamichi, you plan to conduct a customer survey. You will talk to 10 customers randomly entering the restaurant and try to determine how satisfied they are with your restaurant. You estimate this measure to have a binomial distribution and the probability that a guest will be satisfied is 85%.

5. Calculate the probability of the number of successes, i.e. the number of guests that will inform you that they are satisfied and create a plot demonstrating these probabilities.

6. What is the probability that 8 or more of the 10 guests with whom you speak will be satisfied?

7. What is the probability that fewer than 8 of the 10 guests with whom you speak will be satisfied?

Retirement, Financial Independence, and Withdrawal Rates 💸

Can a portfolio containing bonds increase the success rate of an orderly retirement?

What is the required withdrawal rate for such a portfolio to succeed?

As riskier securities, stocks may provide more returns and higher sustainable withdrawal rates, however at what cost, i.e. risk?

What may be the composition of a portfolio that would allow you to withdraw at 4% annually, $3500 each month for the next 20 years?

As assets that are more dependent on the current fiat currency rules continue to struggle, the success rate of such portfolios will dwindle.
In this article, I expand on how retirement funds are constructed and why many of them will fail eventually given today’s gloomy conditions while also answering each question mentioned above. In addition, I will also share my thoughts on what I myself do to survive and explain the illusion of measuring your returns in terms of just percentages and fiat currencies.

Solution to Management Science Series #202: Fundamentals of Sampling in Business Context: Kyoto Games

Kyoto Games decided to work with Selim Hasagasioglu Consulting on the launch of its new hybrid gaming console.

The consultancy firm needs to determine whether the customers Kyoto Games is targeting have a different average price preference from that of the overall market.

The consulting firm randomly samples 45 people from the Kyoto Games’ target market. It is known that the mean hardware price for the overall console market is $500 and the standard deviation is $125.

1.Estimate the distribution of the sample means, explaining your rationale.

2.Do you need to know the distribution of the pricing with respect to overall hardware market and with respect to the sample means?

3.What is the expected value and the standard deviation of the sample mean, which is the standard error? Assume that population is so large that you can ignore the finite population correction factor.

4.Assume that the average price for your sample is $465. What is the probability of getting this value or lower if the target market and overall market have the same price preferences?

5.Assume that the average price for your sample is $540. What is the probability of getting this value or higher if the target market and overall market have the same average price preference?

6.Assume that your sample size is actually 120 people and the average price for the sample remains to be $540. What is the probability of getting this value or higher if the target market and overall market have the same average price preference?

7.Your sample size is 250 people and the average price for the sample stays at $540. What is the probability of getting this value or higher if the target market and overall market have the same average price preference?

Solution to Management Science Series #201: One of the building blocks of data analytics: Probability in the Context of Medicine (Hard)

Based on the mother’s demographics, the probability of a baby having a very rare disease called ‘Rare Disease’ is 0.01.

Prenatal test for the Rare Disease is 98% accurate when the baby has the disease and when the baby does not have the disease.

The mother is tested and the test is positive.

Calculate the probability that the baby does not have the disease given that the test is positive.

Solution to Management Science Series #200: One of the building blocks of data analytics: Probability in the Context of Business (Medium)

One of the building blocks of data analytics: Probability in the Context of Business (Medium)

The probability that a customer buys your streaming service, Halo, is 30%.

The probability that a customer buys your streaming service Halo and signs up for your company’s specific loyalty program is 20%. Every other streaming service offers loyalty program.

Loyalty programs differ among service providers. In order to sign up for the loyalty program, a customer must purchase the streaming service first.

a) Given that a customer buys your streaming service, what is the probability that the customer also signs up for your company’s loyalty program?

b) What is the probability that all customers using all kinds of streaming services also sign up for the loyalty programs of these companies if signing up for the loyalty program is independent of the streaming service they subscribed to?

PS: Current AI tools could not solve this problem.

On why statistics, beefed up as ‘Data Analytics’, is an essential tool to model, predict, and solve business problems

On this planet, understanding the nature of a random experiment solves more than half of the problem at hand.

However, this may be a hard task in some particular circumstances and boundaries are blurred.

How many businesses went bankrupt in the past hour is a discrete variable you can count without trying to predict anything.

How many e-commerce companies will record positive operational cash flows next year may be relatively well predicted through a continuous distribution model.

We live in a world that is more continuous than is discrete. In this article, I expand on this fact, which alone requires everyone involved in the world of business (basically almost everyone living on earth) to know, understand, and use statistics and its tools.

Solution to Management Science Series #199: One of the building blocks of data analytics: Probability in the Context of Business (Easy)

One of the building blocks of data analytics: Probability in the Context of Business (Easy)

The board members of the company, Universe, met to share their forecasts and estimates for the company’s future profit and revenue targets.

60% of the board members think that Universe will beat profit targets while 70% believe that Universe will beat sales.

20% believe that Universe will beat neither profits nor revenue targets.

a)What is the probability that a board member believes that Universe will beat profit, revenue targets or both?

b)What is the probability that a board member believes that Universe will beat both profit and revenue targets?

Historical Returns: Commodity Futures vs Equities. Revisiting the work of Gorton and Rouwenhorst

They did not teach you this in your finance classes 💡

⬇️

When using Treasury bills as the margin collateral for the futures position without leverage, fully collateralized commodity futures have essentially the same returns and Sharpe ratio as equities.

Agree or disagree?
In this article, I will investigate how the returns on commodity futures compare with those on Equities.

Solution to Management Science Series #198: Using K-Means Clustering for Customer Segmentation

Using K-Means Clustering for Customer Segmentation
Even Excel, albeit sometimes clumsy, can be a data analytics tool.

Try to solve the following question by using Excel:

Wine retailer, Porto, had 32 different promotion campaigns. You have the information on promotion month, minimum quantity to be purchased to benefit from promotions, discount percentages, varietals, wine’s origin, and whether the wine has passed its peak or not.

All this information is captured by a single Excel spreadsheet.

Then, you also have customer transaction data in a separate spreadsheet. You see that there had been 324 transactions completed.

Segment customers by using k-means clustering method. Use Excel to do so!