Project 2
Our project 2 is and individual project that uses a secondary data to perform a case study. The topic that i decide to study at is the customer and the shopping power in India because i want to know what is the factors that affect the shopping power in India. I used the dataset from Kaggle ( https://www.kaggle.com/bikram1505/shopping-data/data# ) to perform the testing .
The inferential statistics i decided to do was :
1- sample Hypothesis Testing
- In hypothesis Testing , i want to investigate whether the population mean of Spending Score is 50.00 .
-the result shows that the population mean of Spending Score is 50.00
Correlation
-Correlation is carried out to find out that is there is any relationship between Customer Age and Spending Score
-the graph and correlation test was undergone ,find out that there is no relationship between Customer Age and Spending Score
Regression
-See the relationship of Annual Income and Spending Score , look at how Annual Income affects Spending Score.
- Independent Variable = Annual Income
- Dependent Variable = Spending Score
Shows Positive Linear Model which have a straight-line relationship.
ŷ = 27.76 + (5.56e-05) x , R2 = 0.3485
Estimated changes of the average Spending Score increases by 0.0000556 as a result for increasing one India Rupee in Annual Income.
Only 34.84% of the variation in Spending score is explained by the Annual Income.
Chi Square test of Independence
-Investigate whether is there any relationship between gender and annual income.
The hypothesis is tested using 5% significance level.
P-value 0.01222 < 0.05 and X2 8.8095 > 5.99, therefore, Reject H0. There is sufficient evidence to conclude that there is a relationship between gender and annual income.
Conclusion
Based on the analysis that we had done for the study of customer and the shopping power, we find out that :
- Population mean of spending score is 50.00.
- No any relation of spending score with age of the customer
- There shows relationship between the annual income of the customer with the spending score. With the increase of one India Rupee, the spending score will increase by 0.0000556.
- Only 34.84% of the variation in Spending score is explained by the Annual Income.
- There is a relationship between gender and annual income.
Reflection
Before having hands on experience on doing Project 2 , I learned about the Inferential Statistics through online classes from Sir Chan Weng Howe . I learned about Hypothesis Testing , Chi-squared test , regression , correlation and much more . Even though I had learn the basics of the inferential statistics , I did not have a good grasp of what it is all about. During the process of processing the secondary data , i felt more relevant and have the first attempt to actually do a real inferential statistic rather than just doing exercises and reading from the slides. During the process , it was difficult to start learning R Coding to process the secondary data.However, with the help of Internet , we can learn anything we want in a more faster way and I managed to process the data in the way i desire.I was amazed that we can discover the relationship of things with a more factual and dependable method. Graphs and analysis were made for the inferential statistics , and it strengthen my R programming skills and analytic skills . It was very worth the time learning new things and try the inferential statistics on my own. I want to thanks especially to Sir Chan that assist me a lot in my project 2 and I am very happy and satisfied for the outcome of the project.