Monthly update - Recovered half the last big loss.
I wrote a bot and gave it 25k to trade
- Thread starter _terminus_
- Start date
Had to widen the P&L fields in the bot as positions are 4-digit at this moment.
CPU heat sink bracket finally arrived from eBay Chinese supplier. Trading computer back in vertical position.
CPU heat sink bracket finally arrived from eBay Chinese supplier. Trading computer back in vertical position.
Last edited:
Monthly update.
August best month, 2-digit return. The new high that IB shows could be a little over the top - I have no idea how they calculate it:
My own calculations are as follows:
There was an error in the last report, in that I had mistakenly assigned some trades to July which would have been closed in August.
Also, the monthly %s should be more consistent now, as they are computed on the exact monthly balance, taking into account deposits and withdrawals.
In August I could have made about 5% more. As you may remember, I had re-enabled the old algo, which opens several small trades during the day, while the new algo holds the positions longer. Since I have a parameter for the maximum number of positions that can be opened across all algos, it happened that the high-frequency algo prevented the slower (more profitable) algo from adding more positions. In other words the two were competing over the maximum number of positions and in so doing the overall return was negatively affected.
So I decided to ditch the old algo definitively. Not without some grief, as it served me well in the beginning.
August best month, 2-digit return. The new high that IB shows could be a little over the top - I have no idea how they calculate it:
My own calculations are as follows:
There was an error in the last report, in that I had mistakenly assigned some trades to July which would have been closed in August.
Also, the monthly %s should be more consistent now, as they are computed on the exact monthly balance, taking into account deposits and withdrawals.
In August I could have made about 5% more. As you may remember, I had re-enabled the old algo, which opens several small trades during the day, while the new algo holds the positions longer. Since I have a parameter for the maximum number of positions that can be opened across all algos, it happened that the high-frequency algo prevented the slower (more profitable) algo from adding more positions. In other words the two were competing over the maximum number of positions and in so doing the overall return was negatively affected.
So I decided to ditch the old algo definitively. Not without some grief, as it served me well in the beginning.
Attachments
Last edited:
Have you thought of putting strategy on collective 2 or something?
Good question. Yes, I have considered that. However, I've read some unfavourable comments about such services, mainly regarding the fact that many of their users are gamblers. You have no control on positions that get opened, size etc. and ultimately there is the possibility that returns could be ruined because of that.
I did offer my system to a couple of local banks and they laughed about it. Not that I expected banks to have that much vision and entrepreneurial spirit, but ok.
All in all if I manage to sustain this kind of returns the system could very well finance itself.
I did offer my system to a couple of local banks and they laughed about it. Not that I expected banks to have that much vision and entrepreneurial spirit, but ok.
All in all if I manage to sustain this kind of returns the system could very well finance itself.
Closed everything out into the opening yesterday, +3339 (+6.66%).
Well more fool them(local banks) , very few , and some with big funds , achieve your sort of returns , people have made billions on far less .
I don't know much about collective2 ( which seems a little more stable than some, at a quick glance) , but I don't think your held / are responsible for actions of users /followers .
Last edited:
I meant that if very large positions are opened - which you have no control on - that could potentially play against the effectiveness of a system and worsen the returns.
Well... I can't really tell. It probably wouldn't harm. On the other hand if you look at the last report I posted, the average % gain per trade is currently 0.19%. So even relatively small movements could affect the system in unpredictable ways.
I'd be glad if other posters expressed their opinion.
I'd be glad if other posters expressed their opinion.