The following is an outline for a simple data analysis project.
Please download the dataset https://s3.amazonaws.com/cc-analytics-datasets/Building_Permits.csv. The provided dataset comes from the City of Raleigh Open Data website and is based upon pending/granted building permits.
Here are the following concepts I have worked on
- Load the data from the provided source via a web request rather than downloading a local copy and loading from disk.
- Review the summary statistics for the included features. o Number of rows and columns in the dataset o Total different types of construction o Mean and median number of stories o Standard deviation for the X and Y coordinates of the permits
- Plot the distributions for each of the following features: Estimated Project Cost and Issue Date Month. Describe the distributions for these fields and explain what insights you might be able to gather.
- A hypothetical executive team is interested in the behavior between Permit Issue Year and Estimated Project Cost, but only for the "New" construction work class, with a construction type of "V B", and with less than 3 stories. Perform a simple regression analysis of this relationship and describe what insights we can gleam from this using success metrics. (Hint: Implement handling for missing values and explain your reasoning.)
- What additional techniques or methodologies could be used to improve the results from the previous step?
