Hello, need some help and guidance on how to start these final project for managerial statistics. Would appreciate any directives!

Info to follow:

The final project will be addressing a research question by analyzing a dataset of your interest. Evaluation is based on (1) A reasonable research question (20%); (2) rigor and logic of the inference methods you used (50%); (3) clarity (figures or tables) (30%). The project should be a write-up (research paper) that contains following components.

Section 1: Introduction. For example, you can discuss:

- What is your research question?

- What is your motivation?

- Why do you care about this question?

- Why others should care about this question?

Section 2: Data Description. You may include one ore more of followings:

- Data sources.

- Describe the variables in the datasets.

- What are the data types?

Section 3: Data Summary. You may include one or more of followings:

- Mean, median, standard deviation, etc.

- Histogram, boxplot, scatterplot, qq-plot, etc.

- Or any other summary statistics or visualizations you think is appropriate.

Section 4: Inferences. Based on your research question, you may do one or more of followings: - Hypothesis testing

-Confidence Interval

- Simple linear regression

- Multiple linear regression with variable selection.

Section 5: Conclusion.

Write a brief summary of your findings. Are the data analysis results support your hypothesis? You may find that the methods we covered in this class is inadequate for your goal. Thus, you could also talk about possible future works.

BELOW IS AN EXAMPLE THAT WAS PROVIDED:

Section 1: Introduction.

- What is your research question?

We obtain the dataset from Kaggle, whichis a sample of the transactions made in a particular retail store on 2018 Black Friday. Our research question is that does female spend the same dollar amount as male do in this retail store on Black Friday?

- What is your motivation?

We were curious about the common myth, women spend more than men. We wonder if there is a gender difference in consumption behavior, especially in one the biggest shopping event, Black Friday. Thus we want to study the relationship between gender difference and purchasing decision making.

- Why do you care about this question?

Because Black Friday is one the most important date in retail sector, and getting more and more data-driven analysis can help to predict customers' behavior.

- Why others should care about this question?

Our research would be helpful to understand the relationship different gender issues to final purchase, guide vendors marketing strategy, as well as predict future consumption trends.

Section 2: Data Description.

- Data sources.

We chose a dataset from Kaggle.com of transaction data made in a retail store on Black Friday.

Link: https://www.kaggle.com/mehdidag/black-friday

- Describe the variables in the datasets.

Variables in the dataset: user ID, product ID, gender, age, occupation, city, stay in current city, marital status, product category, and final purchase in dollar amount. We mainly focus on gender and final purchase in dollar amount.

- What are the data types?

Data in the columns of user ID, product ID, gender, age, occupation, city, stay in current city, marital status, and product category are qualitative ordinal data; whereas final purchase in dollar amount is quantitative discrete data.

Originally, we wanted to use multiple variables regression find a more defined pattern in consumer behavior in this particular retailer store on Black Friday. But due to the qualitative nature of our dataset, we can only utilize hypothesis testing and confidence interval. We still believe the finding would have statistically meanings.

Section 3: Data Summary.

The total of transaction is more than 500,000. If we analyze the data, we can see the total number of total sampled customers= 5891

If we segregate the data in terms of gender, we get the following count:

Male= 4225

Female=1666

And total purchase value= 5,017,668,378

Purchase value by males=3,853,044,357

Purchase value by females=1,164,624,021

So, we can see out of total customers, 71.7% are males and they make 76.8% of the total purchase.

-Mean.

Mean of overall purchase value= 851,751.55 (851,752 approx.)

Now let's consider mean as per gender since our research question is based on that.

Mean purchase value of males=911,963.16 (911,964 approx.)

Mean purchase value of females= 699,054.03 (699,055 approx.)

Here we can see the sample mean purchase value of males is 23% higher than the purchase value of females.

-Median.

Median of overall purchase value= 512,612

And let's compare the value in terms of gender again.

Median purchase value of males= 565,925

Median purchase value of females= 398,178

Mean of overall purchase value is greater than the median, that means the distribution is skewed to the right. This distribution seems skewed to the right regardless of gender.

-Standard Deviation.

Standard Deviation of overall purchase value= 932,997.81

Standard Deviation of purchase value for males= 975,397.76

Standard Deviation of purchase value for females= 795,738.05

From these estimations, we can see that males do have a higher standard deviation which means the data points for the male customers are spread out over a large range of purchase values than those for females.

-Histogram.

For detailed understanding of the observations from the mean, median and standard deviation, let's have a look at the histogram gender-wise.

In male purchase histogram, we can observe that most of the values are within the range of 100,000 and 5000,000.

\fFemale Purchase Histogram 1000 900 800 700 600 500 400 300 200 100 40000 100000 500000 1000000 3000000 5000000 7000000 OFrequency 0 82 873 369 297 41 A