How backtesting works and what software is required?

V

vk60546

sorry for a noob question, but can someone explain how backtesting is done? What would be helpful is any strategy (does not need to be profitable for this example) explained and then back tested.

Thanks in advance.
 
A fairly inexpensive way to get started is to go here: http://www.amibroker.com/ (has a 30 day free trial too)

Also, you can use this book to help try out a few systems:
It helps to have a programming background, but not necessary with some work.

Good luck.
 
run you strategy over some data. you can get history from quandl.

here's a world class winner. if the previous day was up: buy the open sell the close
Code: 
import pandas as pd
import numpy as np
import Quandl
df=Quandl.Quandl.get('GOOG/NYSE_SPY',authtoken=QUANDLTOKEN,sort_order='desc')
df['upday']= np.sign(df.Close-df.Open)
df['prevupday']= df.upday.shift(-1)
df['range']= df.Close-df.Open
df[df.prevupday>0].range.cumsum()
 
Option_Attack said:
A fairly inexpensive way to get started is to go here: http://www.amibroker.com/ (has a 30 day free trial too)

Also, you can use this book to help try out a few systems:
It helps to have a programming background, but not necessary with some work.

Good luck.
More...
+1
There is also an active user group with over 14000 users :
https://groups.yahoo.com/neo/groups/amibroker/info
One can also participate without being a yahoo user, by simply subscribing via email (--> mailinglist). See above link for details.
 
vk60546 said:
sorry for a noob question, but can someone explain how backtesting is done? What would be helpful is any strategy (does not need to be profitable for this example) explained and then back tested.

Thanks in advance.
More...
The following MovingAverage Crossover system is maybe one of the simplest systems one can design (it is from the AmiBroker examples that is part of the software):
Code: 
period = 20; // number of averaging periods
m = MA( Close, period ); // simple moving average
Buy = Cross( Close, m ); // buy when close crosses ABOVE moving average
Sell = Cross( m, Close ); // sell when closes crosses BELOW moving average
And the result one can inspect on the chart (with green arrow signals indicating "buy" and red arrow "sell").
And one can get the result also in a spreadsheet table with all the details like date, price, action etc...
And there is a 3rd method of real backtesting with many statistical metrics...
I would suggest to research the many texts and examples on the web...

As already was said it has also an evaluation version for trying out.
But be aware that you have to invest some time in setting up the database of ticker symbols that you are interested in,
and learning the software, as is normal with any new software... IMHO half a day or so...

It already has a demo database integrated for quickly trying out all the features...
 
Last edited:
Amibroker is an excellent choice. Try some examples first and ask questions in forums. Backtesting is a tricky subject.
 
With all the free demo's available, why back test? Your psychological mind is not in the past, but the present. Forward test and challenge your emotions.

Na zdrowie,
Tim
 
You must log in or register to reply here.
Back
Top