|
c*********e 发帖数: 2340 | 2 图是从上方拍的,车库那一角的地面很明显下斜,裂口就在地面最convex的地方。 |
|
g*****9 发帖数: 4125 | 3 My understanding is they are the same thing. I call them pruning shears.
They are strong enough to prune hard branches of trees and shrubs, sometimes
up to two centimetres thick.
There are three blade designs of pruning shears: anvil, bypass and parrot-
beak. Anvil pruners have only one blade which closes onto a flat surface.
They tend to crush the stem, but remain reliable when slightly blunt. Anvil
shears are useful for cutting thick branches. Bypass shears usually work
exactly like a pair of ... 阅读全帖 |
|
r****e 发帖数: 867 | 4 【 以下文字转载自 PDA 讨论区 】
发信人: rotone (Narcissist), 信区: PDA
标 题: iPhone 5S会有指纹识别吗?
发信站: BBS 未名空间站 (Sun Aug 11 22:36:44 2013, 美东)
现在9月10日发布新iPhone谣言满天飞, 这次肯定是5S, 屏幕4寸, 其他的很难搞出太多
花样.
如果能出个home键指纹识别, 跟NFC芯片结合起来, 手机支付感觉会方便不少. 网上密
码 APP密码之类的也可以用这个功能.
就是不知道现在指纹识别技术到底搞得怎么样了. 当年用过IBM笔记本上的, 印象一般,
用它管钱不能让人放心.
MacRumors上说这个sensor只能做成convex的, 不知道真的假的?
http://www.mitbbs.com/article_t/Apple/31678127.html
对iPhone 5S感兴趣的同学去果版押注fingerprint sensor的菠菜吧. |
|
t*******r 发帖数: 22634 | 5 其实那些都是 ppt 用的,说白了一个词 n-polytope 就够了 。。。如果一定要加一个
限制语,那么。。。convex 否?。。。 |
|
t*******r 发帖数: 22634 | 6 更基本?可考性更强?convex & closed form?。。。当然我不搞统计。。。
不过不是最热的那些领域,教育界跟工业界脱节还是挺严重的。。。我们这边都用
nonlinear 算法好多年了,教育界还是在大部头教 linear 算法。。。有人开玩笑说关
键是人教授也不会。。。 |
|
t*******r 发帖数: 22634 | 7 NLP 适合变量多但是 cost function 简单,能够 smooth 做成 continues
differentiable,并且有一个主要 cost func 最好是 convex 的情况。。。如果 cost
function 复杂或者解空间太 fragmented,不适合 NLP。
但你 7, 8 个变量这种七八个星星天外的问题,基本不太会是 NLP 的 sweet spot。。
。我们大多数的实际问题,一般要 1 million 变量才出门跟人打招呼。。。 |
|
t*******r 发帖数: 22634 | 8 再升级到 AMC12,对于 m*n 格子(m >= n),问:
(1)有多少个样子不坑爹的梯形(非长方形,非平行四边形,长底两端都是锐角)。
(2)有多少个 convex 多边形,并且该多边形长得有点像北朝鲜风格的炒鸡蛋。
我不算了,都是特么体力活。// run |
|
t******l 发帖数: 10908 | 9 y = x^2 ,这个得教。也就是 perceptual-conceptual priming。
(对称性/单边之类的 Concrete Operation 分析也略去了)。
在概略画图后,曲线上取两点 (x1, y1) 和 (x2, y2),画一直角三角形,求 slope (
rise-over-run)。(perceptual-priming)。得 slope
m = (y2 - y1) / (x2 - x1)
substitution (semantic priming) 后,
m = (x2^2 - x1^2) / (x2 - x1)
但是我们要 infinitesimal close,所以问题是能不能避免分母是零,或者干脆
cancel 掉分母。这样就 difference of two squares (play with algebraic squares
的一种),得:
m = ((x2 + x1) * (x2 - x1)) / (x2 - x1)
现在是因为 infinitesimal close, 所以没有被零除的问题。
m = x2 + x1
还是因为 infin... 阅读全帖 |
|
|
k********8 发帖数: 7948 | 11 呵呵
以前我很入迷,学校里学了一大堆的这方面的课
什么duration convexity arbitrage binomial tree blah blah blah
现在我就纯粹的一个chart trader了呵呵
我倒没有去算那个coupon payment什么的
in mind that when calculating bond performance, it's not just bond price,
but also need to add bond coupon payments.
Like in the 1970's. |
|
j***b 发帖数: 5901 | 12 Puts are more expensive because the specific way that market index moves.
You see the pattern of segments of convex curves. This shape represents
peoples fear of risk. When there is any sign of risk, people tend to run,
causing the steep downward movement.
Any method of hedging the downward risk will be valued by investors, who
share the dislike of risk. Puts are a great way of hedging, and thus they
are expensive. Similarly, vix future, options are also expensive because
they can also be used a... 阅读全帖 |
|
s*******e 发帖数: 432 | 13 If you do not exercise American put the following can happend
There is a possibility that the intrinsic value of the put could be equal or
large than the put option price. When the the intrinsic value is large than
the option price, you can buy the stock ,exercise the option, use the money
you get to buy another same option but with some profit, so you can profit
without risk. Also if you do not excercise,the other party who sold the
option will be able to take some riskless money off the trade... 阅读全帖 |
|
k********8 发帖数: 7948 | 14 没事可以去学学fixed income securities里面的duration and convexity arbitrage. |
|
a*****e 发帖数: 1717 | 15 Easy to imagine:
delta neutral at the open, convexity |
|
u********e 发帖数: 4950 | 16 ☆─────────────────────────────────────☆
jhsph07 (银杏 - 愚公) 于 (Wed Apr 20 16:19:07 2011, 美东) 提到:
熊市里, 靠扑都贵 (high VIX )。
☆─────────────────────────────────────☆
aripple (aripple) 于 (Wed Apr 20 16:19:32 2011, 美东) 提到:
who told you that....
normally put is cheaper.
theta is irrelevant here,
theta is about dynamics.
your average cost is static.
☆─────────────────────────────────────☆
fullmargin (白浪逐沙滩) 于 (Wed Apr 20 16:19:38 2011, 美东) 提到:
不对,性浪的CALL都贵很多。
☆───────────────────────────────... 阅读全帖 |
|
j***b 发帖数: 5901 | 17 That's just one aspect. Another reason is that market index typically has
the convex shape, that means when it goes up, it goes up ever slower, but
when it goes down, it does so ever faster. |
|
a**d 发帖数: 55 | 18 It's not an arbitrage. The underlying of VXX is not the Oct future. It is
the vix future with constant maturity (1m). The reason for its jump more
than Oct future contract is because the convexity of the vix future curve. |
|
V********e 发帖数: 578 | 19 是不是熊套不是主要问题,关键是你为什么被套住了。难道从下套那天起,股市是
convex向上的?
现在是做空的最好时机,因为你知道股市在高位,高点short了,成功率很高,不要太
贪。股市狂泻了,就没有这么低的risk/reward ratio了. |
|
r****e 发帖数: 867 | 20 【 以下文字转载自 PDA 讨论区 】
发信人: rotone (Narcissist), 信区: PDA
标 题: iPhone 5S会有指纹识别吗?
发信站: BBS 未名空间站 (Sun Aug 11 22:36:44 2013, 美东)
现在9月10日发布新iPhone谣言满天飞, 这次肯定是5S, 屏幕4寸, 其他的很难搞出太多
花样.
如果能出个home键指纹识别, 跟NFC芯片结合起来, 手机支付感觉会方便不少. 网上密
码 APP密码之类的也可以用这个功能.
就是不知道现在指纹识别技术到底搞得怎么样了. 当年用过IBM笔记本上的, 印象一般,
用它管钱不能让人放心.
MacRumors上说这个sensor只能做成convex的, 不知道真的假的?
http://www.mitbbs.com/article_t/Apple/31678127.html
对iPhone 5S感兴趣的同学去果版押注fingerprint sensor的菠菜吧. |
|
c*******y 发帖数: 1630 | 21 这个能这么理解么???
你买MBS的时候是short OAS的吧。顶多赚,carry个OAS delta。
我的理解carry是fixcoupon-COF-costofhedge
OAS是决定book p&l on price
不过你上面说的那些,搁一年前都成立。那个时候book value 30多,
duration gap也极小,convexity也hedge了。
swaption |
|
s***m 发帖数: 6197 | 22 【 以下文字转载自 Quant 讨论区 】
发信人: throwaway (专门注册), 信区: Quant
标 题: 答谢文,谈谈VIX 和 VIX future
关键字: VIX
发信站: BBS 未名空间站 (Sun May 11 18:02:53 2014, 美东)
上次发文求建议,收到很多真诚的帮助。想了想可以写篇我对 VIX index 的学习心得
,算是表达感谢吧。
说明一下,我毕竟是个菜鸟新人,这篇也只是综合公开资料再加上我自己对 VIX 的理
解,也就希望对准备学习这个领域的人有所帮助,熟悉volatility trading的内行看了
肯定觉得简单,有不对的地方请多多指正。
废话完毕。
1. Rationale behind VIX
想了解 VIX index 计算方法的依据,必读的文章是
“More than you wanted to know about variance swap".
http://elis.sigmath.es.osaka-u.ac.jp/research/gs-volatility_swa
具体的证明比较繁琐,我总结就是:
In a n... 阅读全帖 |
|
t***l 发帖数: 3644 | 23 扯吧。futures和underlying index就差个小小的convexity adj,当然刨去dividend后
的意思,还有个P的carry? |
|
发帖数: 1 | 24 如果英镑大幅上涨,FTSE也会立即屁啊叽。FTSE现在是奇葩股市,一个月来呈现直线上
涨趋势,图形不convex,不concave,不S型,标标准准一条直线。1h线看,是一条极窄
的上升通道,非常有喜感。 |
|
l*****n 发帖数: 328 | 25 当那些华尔街的高管,senior traders, managing directors成为公众的众矢之的,坐
在国会听证会上被人骂。他们会不会觉得自己很没面子? 我看不会,他们心里其实得
意的很,风光的很,你不到一定层次,谁会来骂你?曹操被称作一代奸雄, 是褒义还是
贬义? 呵呵。。。。 有人看到过 IT 工程师被人骂吗?
骂他们的人,其实大部分心里还是暗暗羡慕他们的,如果有一个机会让他们去华尔街去
和那些被骂的人一起在trading floor上叱诧风云 ,十个有九个是愿意的。。。。
yield curve, duration, convexity, basis points, long, short, private equity,
morgan stanley, citigroup, goldman sachs, even lehman brothers and bear
stearns. 这些词充满了魔力。 那些traders, 前面4台电脑屏幕,一个bloomberg
terminal, 很多不断滚动的数字,一手一个电话,正前方一个CNBC大屏幕,看着就让人
觉得神秘,让人 |
|
c***z 发帖数: 6348 | 26 Because your talent growth is convex and the company expect that to be
concave, at one point you need to jump ship to adjust for that. |
|
o**********e 发帖数: 18403 | 27 【 以下文字转载自 WaterWorld 讨论区 】
发信人: jinkela1990 (通信WSN), 信区: WaterWorld
标 题: 求助,先接受offer再拒,被亚马逊追杀到学校请喝茶严重么?
关键字: 接受再拒绝 offer
发信站: BBS 未名空间站 (Sat Jun 7 14:43:07 2014, 美东)
小弟是南加某校新毕业EE硕士,5月份拿到A的offer,当时A催的紧不给我Extension,
不得
已5月14号先接下Offer。
我5月20号拿到Q的offer,然后拒绝了接下的A的offer。当时A还威胁我要上告到学
院让我吃不了兜着走。当时仔细看了看条款,是employment at will的,也没在意,就
很客气的感谢了HR的工作。然后跟Q走了。
结果昨天学院院长给我发了个邮件说A告到学院,说我的行为是breach of
professional ethnics,院长要找我下周一喝茶,并且把所有的相关邮件收集好,并且
院长的邮件口气非常硬。小弟直接吓尿了。
在这里求助,咨询一下这种情况下,学院最高会给到什么处罚?影响会很大么?我还有
一周就毕业... 阅读全帖 |
|
c***z 发帖数: 6348 | 28 谢谢大牛的建议,现在我也在考虑跳槽过于频繁的可能副作用了
不过应该关系不大。而且这是迫不得已的,公司假设你的成长是concave的,可是你实
际的成长是convex的,所以必须频繁跳才能得到新的成长机会。
我也有思考大牛说的credit/reputation system,这其实是中世纪的东西,所以很容易
被烙印crack,而且最终会变成妨碍生产力的network system: it is about who you
know, not about what you know
选择只有两个:学习烙印crack这个系统;突破这个系统,建设真正的meritocracy
暂时似乎只能选择1,就是crack the system and make every Chinese reputable/
hirable
希望招人的大牛们能考虑这点,这也是brand image |
|
d**y 发帖数: 32 | 29 两篇文章,请方向相符及有兴趣的发站内信,并请附上简要个人介绍(学历、研究方向
)及
email联系方式,谢谢。
1: Stochastic Control Mechanism on Routing Multiclass Switching Network
Through Game theory
ABSTRACT
Over the last few decades stream control mechanism has more attention in the
early stages of research in communication network. The motivation of this
work is deep understanding on the stream control mechanism problem in multi-
class networks. In this paper, game theoretic perspectives are presented and
the appropriate frame work for the stu... 阅读全帖 |
|
q****g 发帖数: 84 | 30 先waive了再说吧
[在 convexity (凸性) 的大作中提到:]
:本人博后
:现在J1签证,有回国两年限制,没办waiver,J1年底到期
:........... |
|
m****s 发帖数: 7397 | 31 Fund Regulatory Assets Net Assets
================================================================
Bridgewater Associates $121,007,400,554 $121,000,000,000*
Millennium Mgmt $119,000,070,000 $13,512,188,000
Citadel Advisors $115,208,973,792 $12,610,000,000
AQR Capital Mgmt $75,642,756,187 $43,500,000,000
D.E. Shaw & Co. $49,315,000,000 $16,500,000,000
Renaissance Technologies $48,980,049,045 $20,000,0... 阅读全帖 |
|
m****s 发帖数: 7397 | 32 Fund Regulatory Assets Net Assets
================================================================
Bridgewater Associates $121,007,400,554 $121,000,000,000*
Millennium Mgmt $119,000,070,000 $13,512,188,000
Citadel Advisors $115,208,973,792 $12,610,000,000
AQR Capital Mgmt $75,642,756,187 $43,500,000,000
D.E. Shaw & Co. $49,315,000,000 $16,500,000,000
Renaissance Technologies $48,980,049,045 $20,000,0... 阅读全帖 |
|
o**********e 发帖数: 18403 | 33 【 以下文字转载自 Working 讨论区 】
发信人: onetiemyshoe (onetiemyshoe), 信区: Working
标 题: 求助,先接受offer再拒,被亚马逊追杀到学校请喝茶严重么? (转载)
发信站: BBS 未名空间站 (Mon Sep 8 11:16:16 2014, 美东)
发信人: jinkela1990 (通信WSN), 信区: WaterWorld
标 题: 求助,先接受offer再拒,被亚马逊追杀到学校请喝茶严重么?
关键字: 接受再拒绝 offer
发信站: BBS 未名空间站 (Sat Jun 7 14:43:07 2014, 美东)
小弟是南加某校新毕业EE硕士,5月份拿到A的offer,当时A催的紧不给我Extension,
不得
已5月14号先接下Offer。
我5月20号拿到Q的offer,然后拒绝了接下的A的offer。当时A还威胁我要上告到学
院让我吃不了兜着走。当时仔细看了看条款,是employment at will的,也没在意,就
很客气的感谢了HR的工作。然后跟Q走了。
结果昨天学院院长给我发了个邮件说A告到学... 阅读全帖 |
|
u****z 发帖数: 43 | 34 10$ 一本。
Fundamentals of Logic Design
Matrix Analysis and Application
Computer Networking
Digital Communications 4th edition
Convex Analysis and Optimization
Adaptive Filter Theory
Signals and Systems
Introductory Econometrics 2e
Money, Banking & Financial Markets
.....
书的封面可以看这个google photos:
https://goo.gl/photos/gkC7xMhCDkTKQoyLA
书在Fremont Ardenwood 区。 |
|
l*****n 发帖数: 328 | 35 当那些华尔街的高管,senior traders, managing directors成为公众的众矢之的,坐
在国会听证会上被人骂。他们会不会觉得自己很没面子? 我看不会,他们心里其实得
意的很,风光的很,你不到一定层次,谁会来骂你?曹操被称作一代奸雄, 是褒义还是
贬义? 呵呵。。。。 有人看到过 IT 工程师被人骂吗?
骂他们的人,其实大部分心里还是暗暗羡慕他们的,如果有一个机会让他们去华尔街去
和那些被骂的人一起在trading floor上叱诧风云 ,十个有九个是愿意的。。。。
yield curve, duration, convexity, basis points, long, short, private equity,
morgan stanley, citigroup, goldman sachs, even lehman brothers and bear
stearns. 这些词充满了魔力。 那些traders, 前面4台电脑屏幕,一个bloomberg
terminal, 很多不断滚动的数字,一手一个电话,正前方一个CNBC大屏幕,看着就让人
觉得神秘,让人 |
|
x*z 发帖数: 694 | 36 Definitely wrong.
infinity is unbounded, not limited, because if a number
>= infinity, then this number is infinity too.
In Euclidean space, a bounded set may contain infinitely
many points, but the convex combination of infinitely
many points is not necessarily bounded.
Even infinity can be measured (measure theory). For
example, countable sets (set of natural numbers, rational
numbers) has Lebesgue measure equal to 0, but the set of
all the
real number between o and 1, which also contains
infi |
|
s***e 发帖数: 5242 | 37 求【最爽之最多种渔获奖】。
这两天分别和两个巴蒂钓到了
pike
smb
lmb
perch
sunfish
trout
crappie
总收获 > 40条,最后我只keep了5条crappie。
有岸钓有船钓,有溪有湖有河,5个钓点,其convex面积 > 2500 平方迈。 |
|
|
m*********a 发帖数: 2000 | 39 i like your "convex area" |
|
|
n*********3 发帖数: 534 | 41 Should be ok, especially yellow mountain, http://www.ymimports.com/
But it is reviewed that YunZi has lead in it.
and I have a set of natural stones from China. The quality does not seem as
good as YunZi, the size and shape has a larger standard deviation than YunZi
.
And I can't find those natural stone weiqi again now (I have not looked hard
enough though). I think they may be more difficult to make than YunZi. If
you find some site, let me know. Maybe I buy a better set.
I don't like double c... 阅读全帖 |
|
|
|
Y*******i 发帖数: 23 | 44 不错的帖子, 很多有用的信息。 BK2是0.25“的厚度, 感觉用起来如何? 我会觉得
太厚了一些。
也喜欢过Fallkniven F1。 Fallkniven是日本做的, 贴得瑞典的牌子。Fallkniven原
来是家小公司, 靠F1打出的名声。 F1借鉴了北欧传统刀puukko的样式。 为了结实耐
用, 用了convex grind, 而不是puukko的scandi grind。 |
|
O**e 发帖数: 246 | 45 Looks like the grinder will be really handy to put a convex edge on a knife. |
|
|
O****e 发帖数: 3290 | 47 Rated #1 in mixed snow by Ski Magazine.
The 2012 Blizzard Bonafide is a super versatile ski. Tip & Tail Rocker,
Slight Camber Underfoot. The all new Flipcore is a revolutionary new rocker
technology. Natural spring, lightweight floatation in powder, while yet
retaining excellent stability are the immediate advantages of Blizzard's new
construction. All of this is made possible through the flipped wood core (
the ski is actually made upside down in the mold), whose downward-facing
convex side for... 阅读全帖 |
|
l******n 发帖数: 11737 | 48 http://uk.psp.ign.com/articles/113/1138879p1.html
We've heard a lot of rumors regarding the PSP 2 but it has been limited
largely to isolated reports regarding specific features; today, however, we'
ve received a comprehensive rundown of the device's features from someone
with direct hands-on experience with the forthcoming handheld.
According to our source, current iterations of the PSP 2 feature dual analog
sliders that are comparable to those found on the existing PSP but are
unique in that t... 阅读全帖 |
|
c**i 发帖数: 6973 | 49 Whitney Vargas, The Sublime Cats of Cartier; Uncovering the artistry behind
the legendary French house's iconic panther. Wall Street Journal Magazine,
Aug 27, 2011
http://online.wsj.com/article/SB10001424053
111903918104576500663156138544.html
Note:
(a) Cartier SA
http://en.wikipedia.org/wiki/Cartier_SA
(Cartier was founded in Paris in 1847 by Louis-François Cartier)
Frech surnames Cartier or Chartier (from Old French charette cart) has
English counterpart: Carter.
(b) The French surname ... 阅读全帖 |
|
t*******r 发帖数: 22634 | 50 “爱恨情仇”这种函数既不是 convex 又不是 concave,基本
不太容易 converge。
万一哪天运气好 converge 了,用屁股想想就知道 converge
在床上某点。
这种破函数实在没啥研究的价值。。。 |
|