趋势明灯 发表于 2009-4-13 22:08

等待
*d:1*

50126365 发表于 2009-4-13 22:12

:*26*:
等待学习中

落尘520 发表于 2009-4-13 23:06

均线回归于抛硬币,离得有点远了吧,总不能用统计1日均价线和5、10等均线两次交叉的间隔日期来找回归概率吧,偶理解不了。股价运行里有庄散博弈的成分,认为因素很重啊

xiongbin861018 发表于 2009-4-13 23:15

搬凳子坐这等了。:*18*:

lsg2006 发表于 2009-4-13 23:18

均线与股价的关系:先说回归均值理论

  回归 ...
问题就出在你前面说的这句话:“假如前100次抛出的结果都是正面朝上”。

chunart 发表于 2009-4-14 02:33

*d:1* *d:1* *d:1* *d:1*

sdjnyzds 发表于 2009-4-14 03:20

学习中*d:1*

ranchgirl 发表于 2009-4-14 05:13

回复 #58 爱庄股份 的帖子

这是概率论中的一个经典案例:
  一枚硬币,正反两面,假如前100次抛出的结果都是正面朝上,那么,第101次,背面朝上的概率是多大(不考虑硬币立起来的极端情况)?
  答案是肯定的:还是50%。

According to The Bayesian probability calculus, your conclusion is incorrect.

I would think the same as your way if I did not work for a statistics/forecasting company (AirChina, China Eastern, China Airlines(Taiwan), ANA(Japan), Lufthanssa(German) etc. etc. are all our clients)

Bayesian forecasting is much more accurate than the traditional statistics forecasting.

For Your Information (FYI) Only...

[ 本帖最后由 ranchgirl 于 2009-4-14 05:28 编辑 ]

ranchgirl 发表于 2009-4-14 05:25

回复 #58 爱庄股份 的帖子

Have your seen the Movie <<21>>, I don't know the Chinese name:mad:

The bad MIT professor asked the Main character a question on the blackboard, that is according to Bayesian probability calculus (kind of, not accurate)



[ 本帖最后由 ranchgirl 于 2009-4-14 05:35 编辑 ]

ranchgirl 发表于 2009-4-14 05:32

回复 #58 爱庄股份 的帖子

The basic concept is something has already happened (一枚硬币,正反两面,假如前100次抛出的结果都是正面朝上) will affect the future prediction(Probability)

Therefore, 答案是否定的:绝对不是50%。

[ 本帖最后由 ranchgirl 于 2009-4-14 05:37 编辑 ]

长在云上的树 发表于 2009-4-14 07:25

LS的朋友在国内的论坛还是讲中文吧。。毕竟年龄稍大些的人英语教育普及的并不好

长在云上的树 发表于 2009-4-14 07:38

原帖由 爱庄股份 于 2009-4-13 20:33 发表 http://bbs.macd.cn/static/image/common/back.gif
感谢淡泊了赚了就跑、非法操作的加分鼓励!
你们这哪是加分,是加我的负担哇!:*27*: 想偷懒都不成了。
----------------------

均线与股价的关系:先说回归均值理论

  回归 ...
1.扔硬币确实是独立事件,但是前100次正+第101次是负,几率不是50%,两次正的几率就已经是0.25了,这个例子要算起来几率是0.0000000......1,几乎不可能发生

2.股市的涨跌不是独立事件,在股市里,收阳倾向于连续收阳,收阴倾向于连续收阴,周月尤其明显,通常几率在60-70%,在单边市甚至到达80%-90%,这个定理很多年前就被统计学家证明过了,大部分理论都是建立在这条定理上的

ranchgirl 发表于 2009-4-14 07:44

扔硬币确实是独立事件
No! Who throws the coin, how does he/she throw? What coin he/she is throwing....

However, the math model does not care what was the reason. History will affect the future. That is the math theory based on.

ranchgirl 发表于 2009-4-14 07:47

回复 #411 长在云上的树 的帖子

LS的朋友在国内的论坛还是讲中文吧。。毕竟年龄稍大些的人英语教育普及的并不好

That is OK, they don't have to read, they can ignore my posts.

If you like, read here:
http://bbs.macd.cn/thread-1613641-1-1.html

长在云上的树 发表于 2009-4-14 08:15

原帖由 ranchgirl 于 2009-4-14 07:44 发表 http://bbs.macd.cn/static/image/common/back.gif

No! Who throws the coin, how does he/she throw? What coin he/she is throwing....

However, the math model does not care what was the reason. History will affect the future. That is the math t ...
这里说的是理论上,如果有人用特殊的手法,或者特殊的硬币之类的来作弊,当然是不能算数的
你要明白每个试验都是有前提的,电影看看还可,不能拿来当依据
前两年选修了statistics和financial modelling,对这些还算知道些皮毛
股市涨跌和硬币本质上是完全不一样的,硬币例子在此并不合适,等楼主回来正楼吧

[ 本帖最后由 长在云上的树 于 2009-4-14 08:22 编辑 ]

坚强1 发表于 2009-4-14 08:32

当自己的买点出现50%错误,尽快的止损
当自己的买点出现50%正确,就让利润多奔跑一会:*29*:
关键是你选的临界点是否会奔跑:*27*:

ranchgirl 发表于 2009-4-14 08:42

回复 #415 长在云上的树 的帖子

The math model does not care what was the reason, what kind of event you are modeling, stock market or throw coins, history will affect the future. That is the math theory based on.

Movie is using the theory, actually in a wrong way:*29*: , not theory is based on movie...

[ 本帖最后由 ranchgirl 于 2009-4-14 08:44 编辑 ]

ranchgirl 发表于 2009-4-14 08:48

回复 #416 坚强1 的帖子

当自己的买点出现50%错误,尽快的止损
当自己的买点出现50%正确,就让利润多奔跑一会
关键是你选的临界点是否会奔跑

您活学活用了 “The cup is half full, or half empty" in an interesting way!

*d:1* *d:1* *d:1*

ranchgirl 发表于 2009-4-14 08:52

回复 #412 长在云上的树 的帖子

We are discussing based on the fact: "一枚硬币,正反两面,假如前100次抛出的结果都是正面朝上".

At least, we assume that is the fact.

If you change the fact, the conclusion, of course, will be different.

However, the Math model, the theory will not change...

That is the beauty of Math...

[ 本帖最后由 ranchgirl 于 2009-4-14 08:55 编辑 ]

火星000 发表于 2009-4-14 08:56

原帖由 另外的一块石头 于 2009-4-12 19:18 发表 http://bbs.macd.cn/static/image/common/back.gif
楼主不用太肉痛,这点核心技术,说白了,就是修改ma均线参数的小东东,

入门级的,哈哈,

楼主文章不错,纯属调侃,不要见怪。

其实文章中,最重要的是象哲人一样坐着,想哲人一样思考的,哲学的结论。 ...
//////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////又有大师出现了接着学习。b:b b:b
页: 11 12 13 14 15 16 17 18 19 20 [21] 22 23 24 25 26 27 28 29 30
查看完整版本: 爱庄股份:自我总结贴(均线对K线的作用)