检验TA信号有效性 vs. 分析TA信号内在运作机理

coolboy

To describe a professional phenomenon/process in layman’s language is not that difficult. The difficult part is to describe it clearly and cleanly so that many people or the targeted audience will be able to understand it and appreciate it, and understand it and appreciate it with a great pleasure. This requires the author to not only have a long-term investment in both the scientific and artistic thinking but also a short-term motivation stimulated by a challenging environment. It is very unusual to meet an unusual professional under an unusual circumstance to write something unusually interesting to read. I must have unusually good luck these days and I am going to buy a lottery ticket now.

Coolboy salutes to Hai’s knowledge, skills and efforts.


P.S.: Coolboy is looking forward to reading more of Hai’s interesting writings.



[ 本帖最后由 coolboy 于 2008-4-28 20:42 编辑 ]

jsl

原帖由 littletiger 于 2008-4-28 19:55 发表 to JSL, although I did have 3 PRL papers on related area, I am not be able to find anything useful that can be directly applied to stocks. am I too dumb? but I will keep thinking any connections b ...

怎么说你呢，小老虎 ，怎么学物理学的跟学数学一样。真对不起爹妈给你的聪明脑袋！哪天给editor打电话把你的PRL降为PLA。

 

We can’t make direct comparisons between a (close to) ideal physics model to a highly subjectively influenced financial market. Let’s think about the different aspects first before any number crunching.

 

Fundamentals form an external field. As the price goes to the fair value range, we can ignore this for simplification.

 

S are investors’ bias*(purchasing/selling power) and vary in term of magnitude. We might need to give up the quantum effect here and allow continuous change or more states at least.

 

J is how much one investor influence another, through chart drawing, comments etc. It’s a bit too complicated, so we can just assume an average is OK.

 

Price is a measure of mean filed.

Volume is a measure of dynamics, not the first level intrinsic parameter.

 

Each support / resistance level is a phase. Can you use this model to explain 发呆的“大圆顶”？

 

By observing price and volume, can you infer what was going on in the system?

 

lucky99

Fibo ratio & Wave? Sometimes the fibo ration match the price pattern perfectly!

xssl

顶！！！！！

littletiger

To JSL, PLA, physics letter A? my first paper about Ising model was published there. haha.. thank you for your explanation, I see the connections now but still have a lot of questions. need time to digest. check this link, this guy's comment on recent market makes a lot of sense. http://web.wenxuecity.com/BBSView.php?SubID=finance&MsgID=1096755

还在发呆

To Jsl #59

 

Good question indeed! You always squeeze more out of me.

 

What I presented is actually the “proliferation model”. The object being proliferated is not essential. Be it virus, information, bearish / bullish sentiment, or even chips, etc., you name it! The whole merit of the proliferation model is: every proliferation will have three phases: introduction, burst, and fade. Likewise, it doesn’t matter whether round bottom / top is driven by information proliferation. As long as it is driven by some sort of proliferation, the discussion is applicable.

 

To Jsl & Littletiger

 

Please forgive me for my total ignorance of Ising model. Nonetheless I feel it would be an interesting topic.  I look forward to reading you guy’s insightful discussion, better in layman’s words.

 

To Coolboy

 

Reading your post brought me equal, if not more pleasure than what my post might have brought to you.

 

Recently my hands were tied by family obligations which freed my mind for wild thinking. I’ll post what my thinking revealed to me when I have chance. Just there is no guarantee I can produce meaningful posts every once a while. May the unusual good luck be with you.

 

To Lucky99

 

Fib numbers / ratio and wave are mysterious. It won’t be at the end of the list, but it certainly will be the hardest in the list.

 

To All

 

Thanks a lot for your support. Please feel free to share your insight!

littletiger

from Ising or phase trasition point of view, the big up/down days around 03/17 should be at critical point. in physics, at critical point, system can switch dramatically between 2 phases. to 还在发呆, can you explain one more time, what is the definition of critical point/state in your model?

还在发呆

To LittleTiger

 

So a critical point is like the Reynolds number that governs the phase change between laminar flow and turbulence? This is different from the “critical state” in my 4-state model. Consider in general the stock market have many states (e.g., overbought [OB], oversold [OS], bullish rally [BR], bearish sell off [BSO], range bounded [RB], within an up / down channel [UC, DC]), and one can come up with a Markov model to describe the (random) transition among those states. So the market could be in overbought [OB] state today, and it has 80% probability to exit overbought state (that means it could be in one of the following states: oversold [OS], bullish rally [BR], bearish sell off [BSO], range bounded [RB], within an up / down channel [UC, DC]) tomorrow, and another 20% probability to stay in overbought [OB] state tomorrow. In normal cases, the number of next possible states is large and probability is evenly distributed among those states, which makes prediction very hard. For example, suppose today the market is in [OB], and from [OB], the market can move to 6 different states (OS, BR, BSO, RB, UC, DC), each with probability 1/6. So what would be tomorrow’s state? Well, it’s hard to say because there are so many possibilities and all of them are equally probable. Those states are called “non-critical state”. Now, suppose some states, those “critical state”, have smaller number of next states and the probability distribution highly favors even fewer states. For example, when the market is oversold and VIX spike, with 90% probability it will rebound the next day. Those “critical states” make prediction much easier. My example may give you impression that “critical state” is equivalent to “critical point”, which are likely to introduce phase change (from oversold to rebound). However, a critical state doesn’t necessary mean that the market is at extreme or a phase change will show up. For another example, if a major resistance is broken, and VIX stays low, then with high probability the current trend will continue. This is also a “critical state”, because it’s easy to predict the future movement (trend continues). However, there is no phase change involved as the trend continues.

ajs

呀，今天才发现这么好的帖子，怎么不继续讨论了呢？可惜啊。。。

xiaobailong

我也顶

laogua

ding

Lionking

One of the best posts I have ever seen!

Feel luck in HT, so many superstars!

COS

whzdf