Applied-Data-Science--Spring-2018

Please note the structure of this repo as outlined in this document, as well, as key aspects of this project.

| -- Src 

	| -- LibraryInstaller.R

	| -- DefineFunctions.R 

| -- data 

	| -- raw 

	| -- processed
		      
  | -- References  

	| -- README.md

##Project outline

Group Member: Mario Saraiva, Lizhizi Cui

Start date: March 01, 2018

End date: May 10, 2018

This project is based on the Kaggle competition on "House Prices: Advanced Regression Techniques".

The data is available at: https://www.kaggle.com/c/house-prices-advanced-regression-techniques.

Outcomes:

1. Executive report with findings, including but not limited to:

• Different predictive models

• The pros and cons of each model

• Reflections and Recommendations

###Phase 0: Project Setup

• Create Repo, folders, and outline tasks.

###Phase 1: Exploratory data analysis.

• Understand how the data is distributed
• Histograms
• Scatter plots
• Produce descriptive statistics / summaries
• Extract important input variables for the analysis
• Identify outliers
• Identity patterns (if any).
• Make a ranked list of important input variables for the analysis
• Have a sense of robustness of conclusions(sample biased)
• Conclusion as to whether individual factors are statistically significant
• Uncertainties for important estimates
• Define the problem/purpose of the project(assumptions)

###Phase 2: Design different models to be tested.

###Phase 3: Test models.

###Phase 4: Compare models

###Phase 5: Compile findings into a report.