I'm new to investing, but am very interested in the workings of the market. Does anyone know of a (preferably free) way to retrieve historical 52-week highs data? Specifically, I would like to analyze the behavior of stocks that appeared on the highs list five to ten years ago.
4
-
try barchart.com– RossSep 15, 2015 at 16:27
-
Historical 52 week high doesn't make any sense. 52-week is a trailing measurement.– littleadvSep 15, 2015 at 16:36
-
1@littleadv Yes, so if I wanted 52-week highs from, say, 2005, they'd be based on prices from 2004 to 2005. What's wrong with that? The idea is to use historical data to run some (casual) simulations on stocks I could have purchased back then.– RationsSep 15, 2015 at 16:52
-
Might have to use excel and a lil bit of work. But will be interesting to know if you a find a site that does it for you. Maybe you could design one ;)– RossSep 15, 2015 at 18:05
Add a comment
|
3 Answers
Be wary of survivorship bias, particularly in free sources. For example, explore Yahoo!'s historic quotes for CPQ, HWP, and HPQ. Yahoo! (and other free sources, and often paid sources) will generally only have historic quotes (or any information) for companies that still exist today, and they may or may not handle symbol changes correctly.
While free sources will be able to answer questions like "what did XYZ's 52-week highs look like 10 years ago" (XYZ some company that still exists today and has been around at least 11 years), if you try to calculate "what have stocks that made 52-week highs 10 years ago done vs those that didn't", your results will be skewed because each set will only include companies that still exist today. Companies that in the meantime have gone bankrupt, been bought, merged with some other company, were delisted for some reason, etc., won't show up, even though you may have bought them had you been investing 10 years ago.
What you're contemplating doing is called backtesting ("If I use some method to pick stocks, what would it have done for me if I'd started using the method 10 years ago?"), and survivorship bias is a big thing to watch out for.
-
Thanks! I'm aware of survivorship bias, but think it would still be instructive to explore the associations between various indicators and stock prices. Forgive me if this is silly, but how do you actually pull up a list of companies that appeared on the 52-week highs list on any given date? If it helps, I just discovered that I can access Mergent, Datastream, and possibly CRSP data thanks to my school.– RationsSep 18, 2015 at 13:30
-
I don't know of anywhere you can get a list of 52-week highs, I think you'd have to calculate them. sandp.com used to offer a list of changes to the index, so you could figure out what the S&P 500 was on a specific date, then with historical data, calculate 52-week highs (or whatever stat you're interested in). Unfortunately sandp.com has been redesigned since I last went there and I can't find historical index constituents, but maybe I just didn't look in the right place.– blmSep 18, 2015 at 17:14
If you mean the Standard and Poor's 500 Index, you can find historical prices in multiple websites.
One of these websites is wsj.com, which also allows you to download the data as a spreadsheet.
I obtain the historical data which I use from Yahoo:
http://finance.yahoo.com/q/hp?s=XLF&a=00&b=01&c=2000&d=08&e=10&f=2015&g=m
