pandas can read data from any SQL databases that support Python data adapters, that respect the Python DB-API. Reading is performed using the pandas.io.sql.read_sql() function and writing to SQL databases using the .to_sql() method of DataFrame.

As an example of writing, the following reads the stock data from msft.csv and aapl.csv. It then makes a connection to a SQLite3 database file. If the file does not exist, it creates it on the fly. It then writes the MSFT data to a table named STOCK_DATA. If the table did not exist, it is created. If it exists, all the data is replaced with the MSFT data. It then appends the AAPL stock data to that table:

In [33]:
   # reference SQLite
   import sqlite3

   # read in the stock data from CSV
   msft = pd.read_csv("data/msft.csv")
   msft["Symbol"]="MSFT"
   aapl = pd.read_csv("data/aapl.csv")
   aapl["Symbol"]="AAPL"

   # create connection
   connection = sqlite3.connect("data/stocks.sqlite")
   # .to_sql...