Stock Ticker Lists for Backtesting

As a private trader, I like to scan the whole US stock market for opportunities each day. I want a backtesting stock list that is representative of my trading style. However, most other backtesting publications are limited to the stocks in the S&P500 or even the SP-500 index itself. For my backtesting, I choose a much wider selection of stocks.

The foundation of my stock list is the top 7500 stocks on NYSE, AMEX, NASDAQ in 1994. This list is somewhat representative of the world markets too by including ADRs (international stocks cross-listed on the US exchanges.)

Unlike most readily available backtesting data, I included stocks that have since been delisted – either failed, merged, taken private. I bought the premium-priced delisted historical data, as well as the regular stock data, from CSI Data. For example, Lehman Bros (LEH) is in my test set even though they have been delisted from the major exchanges and relegated to the pink sheets with TWA airlines. In the process, both these tickers disappeared from most historical data sets which leaves the remaining data with a strong survivorship bias. Using delisted data more accurately models the markets.

I spent several weeks validating which reduced the test set of stock tickers by screening out dirty data. Of course, this also increased the accuracy.

The list further shrinks as time moves forward and the delisted stocks drop off. I did not purchase new listings because the test set is still more than big enough to make a statistically valid sample. Sometimes new listings appear on their own when tickers get “recycled” – new companies adopt the ticker of delisted companies. At times, especially in the longer test periods, there are actually two different companies using the same ticker (they start with different tickers, one dies, the other changes to the dead company’s ticker and that ticker is applied to all historical data for that stock). These cases are denoted by appending “DUP” to the ticker for one of the companies.

My stock lists are available here, by test period:

May 1994 – April 2004

May 2004 – April 2007

May 2007 – April 2008