日间自动交易讨论区

酒醉今朝

More than half of day trading hours passed, SVXY model stays quietly, no any buying signal so far.

酒醉今朝

Market BUY SVXY Fri Oct 7 13:30:16 2016 at $74.58

酒醉今朝

Market SELL SVXY Fri Oct 7 15:47:16 2016 at $74.86 Result:0.37

酒醉今朝

酒醉今朝 发表于 2016-10-7 02:47 PM
Market SELL SVXY Fri Oct 7 15:47:16 2016 at $74.86 Result:0.37

+0.37%: Have a good Columbus weekend.

酒醉今朝

上个月底，有一则新闻：　“SVXY: Large Outflows Detected ... an approximate $29.9 million dollar outflow..."
这意味着什么呢？请关心SVXY交易的朋友帮助分析、解读。谢谢。

酒醉今朝

酒醉今朝 发表于 2016-10-9 10:35 AM
上个月底，有一则新闻：　“SVXY: Large Outflows Detected ... an approximate $29.9 million dollar outf ...

有网友评论说“For now, SVXY has a big chance to go south.”

酒醉今朝

Market BUY SVXY Mon Oct 10 11:53:16 2016 at $76.28

酒醉今朝

Market SELL SVXY Mon Oct 10 15:47:16 2016 at $76.57 Result:0.38

Hutong9999

酒醉今朝 发表于 2016-10-9 10:35 AM
上个月底，有一则新闻：　“SVXY: Large Outflows Detected ... an approximate $29.9 million dollar outf ...

The underlying of SVXY is VIX futures. There are also quite a few ETF/ETN tracking VIX futures. I don't think we should read much into its money flow.

Hutong9999

酒醉今朝 发表于 2016-10-10 07:19 AM
有网友评论说“For now, SVXY has a big chance to go south.”

I agree with his statement "The outflow of $29 million does not mean much." He did not explain why has a big chance to go south. VIX is in the middle right now, it can go either way. It is range bounded at the moment and I assume it stay bounded before breaking the range.

酒醉今朝

Hutong9999 发表于 2016-10-10 04:21 PM
I agree with his statement "The outflow of $29 million does not mean much." He did not explain why ...

Thanks for the comments. The volume is low, and the price stays in ranged, which makes SVXY model not perform well as last month.

littletiger

酒醉今朝 发表于 2016-10-10 07:17 PM
Thanks for the comments. The volume is low, and the price stays in ranged, which makes SVXY model  ...

does your algorithm take volume into consideration? From what you described, it only uses price.

any trade today?

酒醉今朝

littletiger 发表于 2016-10-11 09:33 AM
does your algorithm take volume into consideration? From what you described, it only uses price.
...

Not take into account the volume right now, which is why I am seeking a way to add it. Any idea? Please share.
So far 3 hours has been passed since Opening, SVXY model has no any buying signal yet.

littletiger

use it as a weighting factor?

酒醉今朝

littletiger 发表于 2016-10-11 11:26 AM
use it as a weighting factor?

Good idea. The following questions are which need to be weighed, and how they should be weighted. Any further thought, please help.

酒醉今朝

Market BUY SVXY Tue Oct 11 12:53:00 2016 at $73.80

littletiger

but without knowing your algorithm structure, i am afraid that it is hard to fit "weighting" idea in your theoretical framework.
can you give me some reference paper that are relevant to your theory? (i do have some physics background :-))

酒醉今朝

Market SELL SVXY Tue Oct 11 15:47:15 2016 at $72.72 Result:-1.46

酒醉今朝

littletiger 发表于 2016-10-11 02:22 PM
but without knowing your algorithm structure, i am afraid that it is hard to fit "weighting" idea in ...

I use Poincare map to simplify and re-sample stock time-price series, which is similar to Chaos attractor analysis.
Then, do convolution, where multiple time scales, at least three scales: long, middle and short, are involved.
I measure violence of oscillation by Root mean square (RMS).
It seems the larger volume, the larger RMS.
No idea on any relation between volume and time scaling, or frequency of oscillation.
It may need more evidence ...

酒醉今朝

littletiger 发表于 2016-10-11 11:26 AM
use it as a weighting factor?

A technical question as you automate your order via IB API:
Do you use "SMART" or "IBALGO" as market order type?
Does "IBALGO" result in less slippage, better than "SMART"?