Adventures in Automation

Does FasterQuant/AlgoTerminal not have the capacity to use 24 hour data or was this a choice you made regarding data collection when you started the project and are now stuck with it?

The last two trades with 250 RTH 30 min bar targets (19 days) and stops approximately 1/3 of the current daily ranges seems to be fit on a larger data set not in touch with current volatility.

19 days in the current environment of erratic trade policies and twittersphere economics are definitely different then most of the last 10 years docile upward grind.

Protecting unrealized profits is a different issue.
FasterQuant and AlgoTerminal both easily support any market time configurations (RTH, ETH or whatever you want to configure). What I would like to be able to do is take my signals via RTH data but be able to exit via ETH data. This is easy to do with live trading but is not easy to do via a backtest.

The trailing stops I had for these trades were 2 X the ATR of last 14 30 minute bars. So the stops were in line with the current volatility and would adjust accordingly to forward volatility.
 
And this folks is why I do not attempt to survive on trading profits.

Exit 1 ES contract at 2887.25
-$229.10

Exit 1 NQ contract 7606.25
+$355.9

At the close of yesterday these two positions were up $6k. You need ice flowing through your veins to play this game.

I feel your pain man. Sometimes the market is just too wild.

For example, my current NQ system barely makes any money in the back test on last Dec 23 and 24 even though those were huge moves down. The price action during those two days was simply unique and anomalous.
 
Back
Top