Final Project
Example descriptive statistics analysis, step by step
This document is just a walk-through example of how data analysis might be done and why. It is NOT directing students to do all of this. It is simply provided to provide a better idea what data analytics is all about, and an example to help you figure out your project.
1. Note what data is on hand. In this case:
a. Income in terms of billions of dollars from last year, 2018. This income is from the accounting department and considered accurate and the latest information available.
2. Note what decisions were made on how to evaluate outcomes, i.e., exactly what numbers will lead to what decision?
a. The manager of Exxon has decided 3% greater average sales in Asian market compared to the European market will mean that he will build a new refinery in Asia.
b. The manager is also concerned with how regular the income will be as well as how much more income. If the variance in income from Asia is much high than Europe, he will NOT build a refinery in Asia, as that income will be unreliable. He has defined excessive variance at 10% difference.
c. The manager is also concerned with how regular the income will be as well as how much more income. If the variance in income from Asia is much high than Europe, he will NOT build a refinery in Asia, as that income will be unreliable. He has defined excessive variance at 10% difference.
3. First look at the difference between measures of central tendency/average:
a. Example:
i. The case study reads thus: The manager of Exxon has decided 3% greater average sales in Asian market compared to the European market will mean that he will build a new refinery in Asia.
ii. Look at the average data to find out what the outcomes are.
1. The Asian sales are 90 billion dollars and the European sales are 80 billion dollars. The difference is divided by the average of both – 10/85 or 12%.
iii. You advise the client to build the refinery in Asia.
4. Second look at the differences between measures of variance/standard deviation:
a. Example:
i. The case study reads thus: The manager is also concerned with how regular the income will be as well as how much more income. If the variance in income from Asia is much high than Europe, he will NOT build a refinery in Asia, as that income will be unreliable. He has defined excessive variance at 10% difference.
ii. Look at the standard deviation data to find out what the outcomes are:
1. The Asian sales standard deviation is 8%, and the European sales standard deviation is 7%. The difference is 1/7.5 or 17% difference.
iii. You advise the client not to build the refinery despite the greater Asian sales because he has stated that he will not go forward with the project if the variance is over 10%.
5. Third look at the distribution, the chart.
a. Example:
i. The case reads thus: The manager wants to know why there is variance over the course of the last year and wants you to track income by quarter.
ii. You analyze a chart of profits by quarter and find that the Asian profits decline in the beginning and end of the year and are very high in the summer and fall.
iii. You advise the client that there is something the reduces profit in Asia in the spring and winter that should be investigated.
6. Conclusion
a. The difference in income between the regions is substantial and suggests building facilities in Asia. Yet, the difference in variance shows that will be risky. Therefore, management should not go forward with building the facilities in Asia unless a different view on how much risk is acceptable is chosen. The facility should not be build. Further investigation into the reasons for the seasonal differences might lead to a different decision, or at least better understanding.
7. The reader is invited to ask further questions.
Profit: European vs. Asian, 2018 (billions)
EuropeanAsian
Average8090
Standard Deviation78