Backtesting: Excel VS Wealthlab for intraday strategies

E

etfarb

I am looking to back test price action on 1 and 5 minute bars for a couple of strategies but I'm not sure whats the best way to go about this.

I was wondering if anyone here uses excel to backtest a super short term strategy. If so how do you go about backtesting price action in excel on 1 and 5 minute bars?

I am also looking at maybe buying wealth lab to do this but i'm not sure its worth the price. Ultimately I like to keep things simple and cheap so if intraday scalping can be backtested in excel, i'd love to hear from you guys!
 
You appear to be just starting ... please consider at least using proper tools if you ever want to have a chance at succeeding.

Requirement #1: lots of historical data for the markets you are considering backtesting - 10+ years is ideal - how many 1min bars does this make for just one symbol, and how does this fit in Excel ?

Requirement #2: if you ever are able to identify through backtesting a profitable strategy, then you should want to trade it using the exact same software - yours & the platform it is running on. I don't think Excel will help you much there.

IMO, you need a platform like NinjaTrader, MultiCharts, etc. plus quality historical data (that must be bought, no doubt about that) to get started. By chance, Ninja is free until the time you are ready to go live (if ever), it is not perfect (no trading platform is) but that would be my recommendation if you can program in C# or are willing to learn.


And forget about the idea of backtesting 1 or 2 strategies over the next couple of weeks and going live making money with that. It is more likely to take you (anyone) a couple of years - or decades - before getting there.
 
Quote from dom993:

You appear to be just starting ... please consider at least using proper tools if you ever want to have a chance at succeeding.

Requirement #1: lots of historical data for the markets you are considering backtesting - 10+ years is ideal - how many 1min bars does this make for just one symbol, and how does this fit in Excel ?

Requirement #2: if you ever are able to identify through backtesting a profitable strategy, then you should want to trade it using the exact same software - yours & the platform it is running on. I don't think Excel will help you much there.

IMO, you need a platform like NinjaTrader, MultiCharts, etc. plus quality historical data (that must be bought, no doubt about that) to get started. By chance, Ninja is free until the time you are ready to go live (if ever), it is not perfect (no trading platform is) but that would be my recommendation if you can program in C# or are willing to learn.


And forget about the idea of backtesting 1 or 2 strategies over the next couple of weeks and going live making money with that. It is more likely to take you (anyone) a couple of years - or decades - before getting there.
More...

Decades? lol
 
Quote from etfarb:

+1 for excel?
More...

I was a long time Amibroker user but went back to using Excel for backtesting. It's entirely flexible and doesn't suffer from the quirks that you'll find in a 3rd party application. I also find it easier to identify bad data using Excel which is inevitable regardless of your supplier.

My testing is almost exclusively short term intraday (tick - 5 min data). The key for me getting reasonable speed was to store data in .csv files and then use text queries to pull them into excel. Obviously it takes a bit of VBA to pull it together, but you want to keep as many calcs as possible in the sheets to take advantage of multithreading.
 
Quote from etfarb:

I am looking to back test price action on 1 and 5 minute bars for a couple of strategies but I'm not sure whats the best way to go about this.

I was wondering if anyone here uses excel to backtest a super short term strategy. If so how do you go about backtesting price action in excel on 1 and 5 minute bars?

I am also looking at maybe buying wealth lab to do this but i'm not sure its worth the price. Ultimately I like to keep things simple and cheap so if intraday scalping can be backtested in excel, i'd love to hear from you guys!
More...

it might depends on how or what exactly are you planning to use in your calculations.
i mean-you don't want to write everything from scratch(like some indicators or whatever) and then spend another month testing your code. some of the common stuff been coded already in many apps like wealth lab(btw- i think you can find some old copy for free )
but if it's just some plain math\stats excel should be more than enough. specially with VBA
 
Excel is not a trading platform, nor is it a backtesting platform.
Tradestation is
Multicharts is

Excel is just an all-purpose calculation tool. You'll really have to do a lot of work in VBA and formulas to pull this off. Also, you'll likely find that you need to integrate with a database of historic data. Good luck with that.
 
You can incorporate excel into studies with Sierra Chart and use it for trading I believe. They have their own spreadsheets to use with the software.
 
Quote from dom993:

You appear to be just starting ... please consider at least using proper tools if you ever want to have a chance at succeeding.

Requirement #1: lots of historical data for the markets you are considering backtesting - 10+ years is ideal - how many 1min bars does this make for just one symbol, and how does this fit in Excel ?

Requirement #2: if you ever are able to identify through backtesting a profitable strategy, then you should want to trade it using the exact same software - yours & the platform it is running on. I don't think Excel will help you much there.

IMO, you need a platform like NinjaTrader, MultiCharts, etc. plus quality historical data (that must be bought, no doubt about that) to get started. By chance, Ninja is free until the time you are ready to go live (if ever), it is not perfect (no trading platform is) but that would be my recommendation if you can program in C# or are willing to learn.

And forget about the idea of backtesting 1 or 2 strategies over the next couple of weeks and going live making money with that. It is more likely to take you (anyone) a couple of years - or decades - before getting there.
More...

Dom,

thanks for the reply and advice but i'm confused on a few things.

Yes i am just starting out in regards to backtesting but if i am backtesting an intraday strategy why would i need data that goes 10+ years out? markets are constantly changing. if i am looking to backtest a basket of financial stocks why should i go further than 2009?

Not really following you on the platform like ninja/or multicharts for the backtest. I have been looking into multicharts but i'm not really sure what backtesting in your platform has to do vs doing it in excel. I thought the whole purpose of backtesting is to develop an idea, test it to determine its statistical significance (if any) and then trade it. Why must this be done on a platform as opposes to excel?

Lastly... I am fully aware that the road to success takes a lot of time but years or decades...?
 
You must log in or register to reply here.
Back
Top