Write a Python script (preferred) or use Excel file to generate an analytical report
Budget: $30 – $250 USD
Write a Python script (preferred) or use Excel file to generate an analytical
report to address following problem.
2. Submit in a group with a .zip file (report in pdf and script file or Excel file)
The case(s) of Smartwater
You’re the regional manager at Amazon, in charge of 350 wholefood market warehouses in the
region, and the next project you’re going to take on is to optimize the weekly re-stocking of
Glaceau Smartwater. How many cases of Smartwater do you need to re-stock every week?
You want to know many cases of Smartwater to order weekly, for each store, so you minimize the
amount of inventory that ends up sitting idle in wholefood shelves.
You know there has to be a better way to get to a plausible answer that doesn’t involve visiting
every single wholefood in your region and get their sales numbers every single week.
You don’t need to know the population distribution. Since you’ve taken a few statistics classes,
the Central Limit Theorem comes to mind. You know that, applied to real-world problems, the
Central Limit Theorem helps you balance the time and cost of collecting all the data you need to
draw conclusions about the population.
You remember the definition of Central Limit Theorem for sample:
When we collect a sufficiently large sample of n independent observations from a population with
mean μ and standard deviation σ, the sampling distribution the sample means will be nearly
normal with mean = μ and standard error = σ/ √n
Do following steps:
1. Create the population dataset by using random generator and save as CSV and saving it
to avoid always recreating the dataset
a. Picked at random. Why do you need random?
b. Representative of the population.
c. Big enough to draw conclusions from, which in statistics is a sample size greater or
equal to 30.
d. Include less than 10% of the population, if you’re sampling without replacement.
Why you need less than 10% ?
2. Read the output CSV in case you've already generated the population dataset
3. Take a different number of samples with size n=30 and 100, all with the same and plot
the sales data just to see how it looks like.
4. Does the distribution of the sample data by itself have the shape of a Normal Distribution?
Explain about the different standard deviation, or the distance from the mean in both
sample size.
5. How close are you to the population mean?
6. Write a conclusion (itemized) as take away for your boss
report to address following problem.
2. Submit in a group with a .zip file (report in pdf and script file or Excel file)
The case(s) of Smartwater
You’re the regional manager at Amazon, in charge of 350 wholefood market warehouses in the
region, and the next project you’re going to take on is to optimize the weekly re-stocking of
Glaceau Smartwater. How many cases of Smartwater do you need to re-stock every week?
You want to know many cases of Smartwater to order weekly, for each store, so you minimize the
amount of inventory that ends up sitting idle in wholefood shelves.
You know there has to be a better way to get to a plausible answer that doesn’t involve visiting
every single wholefood in your region and get their sales numbers every single week.
You don’t need to know the population distribution. Since you’ve taken a few statistics classes,
the Central Limit Theorem comes to mind. You know that, applied to real-world problems, the
Central Limit Theorem helps you balance the time and cost of collecting all the data you need to
draw conclusions about the population.
You remember the definition of Central Limit Theorem for sample:
When we collect a sufficiently large sample of n independent observations from a population with
mean μ and standard deviation σ, the sampling distribution the sample means will be nearly
normal with mean = μ and standard error = σ/ √n
Do following steps:
1. Create the population dataset by using random generator and save as CSV and saving it
to avoid always recreating the dataset
a. Picked at random. Why do you need random?
b. Representative of the population.
c. Big enough to draw conclusions from, which in statistics is a sample size greater or
equal to 30.
d. Include less than 10% of the population, if you’re sampling without replacement.
Why you need less than 10% ?
2. Read the output CSV in case you've already generated the population dataset
3. Take a different number of samples with size n=30 and 100, all with the same and plot
the sales data just to see how it looks like.
4. Does the distribution of the sample data by itself have the shape of a Normal Distribution?
Explain about the different standard deviation, or the distance from the mean in both
sample size.
5. How close are you to the population mean?
6. Write a conclusion (itemized) as take away for your boss