STATGU4205

Linear Regression Models

Study Notes

1. Why Poisson? Why Exponential?
2. Why MLE?

Extra 5%

1. Data Source: New York City Department of Finance Annualized Rolling Sales

2. Raw Data (around 340 mb) are arranged and stored here

3. Cleandata (around 25 mb) and data description are provided here

   Anyone with a valid columbia email is authorized to download.
   clean_sales_data.m is the script I wrote to clean the data.
   It takes around 40 minutes to run the script on my laptop.

4. More analysis will be published here if time allows.