This is a study project using fictional data from Kaggle, available in http://kaggle.com/harlfoxem/housesalesprediction
- Created this using python for data science, streamlit and Heroku
- Cool dashboards available in:
House Rocket is a real estate company that has started its operations in the northwestern US.
Currently, the business team can't make good decisions without analysing the data available. Also, the portfolio is huge, and it takes a lot of work and time to analyse it manually.
The mission here is to serve the teams with valuable insights in a more dynamic waya - Which homes should House Rocket buy and what should they pay?
b - Once a home is bought, when is the best time to sell it and what should they charge?
- All homes in the dataset are available for purchase by House Rocket.
- Only homes above "condition 3" (i.e. good conditions) to be considered.
- The Sale price calculation suposes that all homes in the dataset are owned by House Rocket.
- Home prices are affected by the season of the year.
- Step 01: Fix datatypes in the dataframe and adjust the data that needs to be adjusted.
- Step 02: Plot the information in maps to visually understand the market distribution over prices and several features.
- Step 03: Organize general market data using tables and graphs, such as price distribution and house aging.
- Step 04: Build insights to define what houses HR should buy, usually houses that are under the price range for a given region.
- Step 05: Set a strategy considering seasonal price changes in the market (hypotetical)
- Step 06: Organize the information in an interactive page with filters using Streamlit
- Step 07: Publish it on Heroku cloud for more democratic access