Search results

I'm basically having the same question as the person who started this thread. I called QP2, but they told me their data does not include delisted securities and does not allow to retrieve the unadjusted (for splits) data.

In my search for historical data that includes delisted securities (to avoid survivorship bias in backtesting), I found that Tradestation is one of the few vendors that offers these data. When talking to them I was under the impression that the data stays on their server however, and that you...

Please find below just one example when "your number" might come up. It's an exaggerated one, I admit, but it seems that's the only way to wake people up. "Before World War I Germany was a prosperous country, with a gold-backed currency, expanding industry, and world leadership in optics...

Michael B, I admire your activity on these forums and once enjoyed myself by reading through your entire journal on your grid systems. I actually simulated the whole thing in VB/Excel, using all sorts of different rules, grid sizes etc. Even before I started these simulations, I concluded...

I can think of only one case when it *could* make sense. This is when the two accounts are with different brokers that somehow differ in the way they fill your orders. In that case some sort of arbitrage could require having both long and short positions in the same currency pair. Since...

I think to be successful you need to be much more sophisticated than just looking at some correlations. this might be of your interest: Pairs Trading: Quantitative Methods and Analysis Ganapathy Vidyamurthy ISBN: 0-471-46067-2 Hardcover 210 pages August 2004

I second that. I do all my backtesting, walk-forward testing, optimization etc. in R, and it works great. You need to get familiar with vectorization though, since loops are very slow compared to e.g. C and Fortran and should be avoided if speed is an issue. I think one of the main advantages...

A while ago, Equis told me it IS actually possible to access all the data fields through DDE, although they don't offer support and you pretty much have to figure it out by yourself. There's a topic on this in the forum: http://forum.equis.com/viewtopic.php?t=3733 I would be interested to...

Metastock with a Quotecenter subscription offers ~23 yrs of fundamental data http://www.equis.com/products/realtime/quotecenter/?overview

I think you also might want to ask yourself the question after how many days (some number between one and never) of trading you're planning to perform another set of backtests, and perhaps calibrate. Then you might want to do a series of walk-forward-tests using all your data (e.g. 4000 bars)...

that's very helpful, thanks. It seems Yahoo does not show delisted securities though, so it's a first step. Eventually, I'm looking for a data source that includes delisted securities as well, in order to avoid survivorship bias.

Many data vendors offer EOD data going back 15-20 yrs. Can anyone recommend a source for EOD securities data going back much longer (1950~), ideally split-adjusted and (most important) including delisted stocks?

Use The Downloader (it comes with Metastock or a subscription to Reuters Datalink) if you're only interested in EOD data. It's a very convenient program, and allows you to update symbols from a library maintained by Equis. I use it to download data for thousands of stocks overnight, and it works...

I'm looking for an inexpensive source for historical intraday split-adjusted equity data that allows me to download historical tick or 1-minute data on a regular basis (daily, weekly) for a large number of equities and save it in ascii (or other common) format. I don't need a realtime datafeed...

Here's my 2 cents: I use a walk-forward approach with two parameters: number of in-sample days (IS) and number of out-of-sample days (OS). Any models/algorithms I build are calibrated on the IS data. I evaluate the system by looking at OS performance, and the ratio of OS/IS performance. Before...

R http://www.r-project.org/

kt, I thought about HDF5 but it seems it doesn't allow indexing through the R api--I understand you either have to load the entire database or nothing at all.

Would using MySQL imply storing the data for each stock in a separate table, and indexing them by date/time? Do you have any experience with the time it takes for a typical query in this case (e.g. 1000 stocks x 10,000 observations).

I use statistical software (e.g. Stata, R) for modelling of returns and large-scale portfolio optimizations. I usually can hold all my (end-of-day) data in memory, but am thinking of using a database engine to facilitate data storage. I might want to start running models on intraday data and...

Last position was stopped out. I'll buy CTHR again at today's close. Stop-loss as before.