Accessing & managing financial data

Exercises:

  1. Read the book chapters Accessing & managing financial data and WRDS, CRSP, and Compustat entirely. If you prefer to read Python code, the chapters are available here and here. They may contain some advanced concepts but also a description of almost every important dataset relevant to research in empirical finance. To get more information on how to compute returns adjusted for delisting, follow the procedure described in Chapter 7.2 of the book Empirical Asset Pricing.

  2. Download the file crsp_monthly.parquet from Absalon. Store it in the folder called data-r (data-python) within your standard working directory for the course. Almost all exercises from now on will start with reading data out of this file, so make sure you familiarize yourself with this short minimal setup to load data into your R or Python session memory from a fresh session (you can consult it anytime again later during the course).

import pandas as pd
pd.read_parquet("data-python/crsp_monthly.parquet")
library(arrow)
crsp_monthly <- read_parquet("data-r/crsp_monthly.parquet")
  1. If you need help with .parquet files, post your question on Absalon. Your TA, your peers, and I will help you!
  2. Replicate the following two figures provided in the lecture slides: i) Create a time series of the number of stocks in the CRSP sample, which are listed on NASDAQ, NYSE, and AMEX. ii) Illustrate the time series of total market values (inflation-adjusted) based on industry classification siccd. The book Empirical Asset Pricing (Bali, Murrey, and Engle) provides a detailed walk-through if you need help.

Solutions: All solutions are provided in the book chapter WRDS, CRSP, and Compustat (R version and Python version) and the lecture slides.