Produce the project based on a transaction data from the company.
What I would want:
- Perform central tendency measures
- Perform spread (variance) measures (one or two)
- Perform correlations, linear regression, time series-forecast Analysis
- I'd also like to apply some data mining techniques like Association rules, Clustering and maybe some third one like Classification
- I will want to use Tableau integrated with R for visualizations
- You will have two datasets in power pivot format to work with (each, 40-45 columns, first one: 5 millions rows, second one: 23000 rows).
Data comes from data warehouse so it doesn't require any special data cleaning except for changing the column names.
As to analysis........I'd like you to focus mostly on data ming: association rules, classification and clustering and add some popular statistical analysis like regression, central tendency and variance measures + perform the 12 month forecast for car insurance claims dataset.