- ago
When backtesting in WL7 we have the option to “Filter Pre/Post Market Data” by checking a box in the strategy settings panel. With this box checked, if a stock price changes significantly during pre and/or post market trading, and we are using a 5-minute scale, won’t indicators with a period of 50, for example, be inaccurate for most of the following trading day because of the omission of the pre and post market activity?

When backtesting in WL*6*, this option didn’t yet exist, and so I simply set it up so that buys and sells could only happen during open market hours from 9:30am to 4:00pm. With this approach, the indicators factored in all available data. Is there a way in WL7 of still using pre and post market data for the calculation of indicators, while constraining the backtest trades to only occur during open hours?

I hope these questions aren't too convoluted. I’m trying to wrap my smooth brain around the complexities of the issue.
1
641
3 Replies

Reply

Bookmark

Sort
- ago
#1
QUOTE:
When backtesting in WL*6*, this option didn’t yet exist

It existed by virtue of the Market Manager's open and close time which acted as natural filters for premarket and aftermarket data. Additionally, the IQFeed provider had its own filter called "Regular Session Only" on its WL6 Data Manager's tab.
0
- ago
#2
Thanks for clearing that up for me. Can you speak at all to the question I raised about the accuracy of the indicators?
1
Glitch8
 ( 8.38% )
- ago
#3
It depends on how you define accurate. The indicators are simply calculations based on the input data. They accurately produce a result based on the input you give them. That's why you have the option to choose to bring in pre/post market data or leave it out. Different traders will was it and others won't. But I've never seen a case of indicators being calculated WITH pre/post market data when the chart is not displaying that data.
1

Reply

Bookmark

Sort