n00b101

31 ms·

241.
by n00b101 12y ago
This is called spurious correlation. It's well known in financial / economic time-series analysis. The lesson is that you never measure the correlation between the PRICE LEVELS of products, instead you measure the correlation betw
242.
by n00b101 12y ago
> the valuation is way north of $1bn... That is an understatement ... MSFT's current market cap is $366 bilion and the all-time high market cap was $616 billion in December 1999.
243.
by n00b101 12y ago
> But if the lender decides to take their stock back from you the borrower, then you are suddenly and unexpectedly in the position of naked short selling? Yes, short sold shares can be recalled at any time. If the shares you short sold a
244.
by n00b101 12y ago
I think the following approach is more appropriate, at least for big business: http://www.rti.com/whitepapers/Dirty_Little_Secret.pdf
245.
by n00b101 12y ago
> How would you overcome the very large inertial barrier that exists in that world to using something new and untested? I've already successfully done this. The sales cycle can admittedly be long, but it's not impossible. You h
246.
by n00b101 12y ago
I've worked for many large financial institutions as a quant and am intimately familiar with use of spreadsheets in critical functions. I believe this is a multi-billion dollar opportunity. If someone is interested in start-up ventures
247.
by n00b101 13y ago
ThinkPad Carbon X1
248.
by n00b101 13y ago
Entrepreneur: I plan to launch a better [XYZ] hosted on Amazon AWS. VC: Well since Amazon has paid all the telcos so that services delivered through AWS "telco-optimized elastic IPs" can be free on data plans, all you have to do
249.
by n00b101 13y ago
Taleb has a good point about people mistakenly interpreting standard deviation (sigma) as Mean Absolute Deviation (MAD). I like that he gives some conversions (sigma ~= 1.25 * MAD, for Normal distribution). I think it's rather silly to
250.
by n00b101 13y ago
This is great, thanks!
251.
by n00b101 13y ago
I know this doesn't need saying ... But I can guarantee you that there are ~25 year old bankers at the investment bank who are making more a lot more than $100k. Technology roles are discounted at investment banks, since they perceive
252.
by n00b101 13y ago
I love my X1 Carbon.
253.
by n00b101 13y ago
Now may be a good time to revisit an old adage of investing: never catch a falling knife.
254.
by n00b101 13y ago
On Wall Street, this kind of rhetoric is called "talking your book."
255.
by n00b101 13y ago
This sounds like a good option for SaaS providers.
256.
by n00b101 13y ago
I think that's a fair point. NVIDIA GPUs are co-processors and need a host system, which adds significantly to the cost, power draw and size. You can try to minimize this (e.g. by maximizing GPU density in a single server, for example
257.
by n00b101 13y ago
The Parallella Gen-1 has single precision throughput of 90 GFLOPS, power draw of 5W, and unit cost of $99. This yields 0.91 FLOPS/s per dollar and 18 FLOPS per Watt. By comparison, the NVIDIA GTX 780 Ti GPU has single precision through
258.
by n00b101 13y ago
This issue goes well beyond "market knowledge." MtGox _is_ the market; they control the order book, trade matching engine, data feeds and even the custodian accounting system. If you own your very own completely unregulated tradin
259.
by n00b101 13y ago
> it will cost you 1 BTC to produce, market, and sell a single widget, and that it is only worth your effort if you can sell each widget for 2 BTC, a handsome 100% margin. That's a 50% profit margin, not 100%. profit margin = 1 - co
260.
by n00b101 13y ago
Do you have any references for the approach that you recommended? You're saying that differentiation (not even just numerical differentiation) is "not something you want to apply to real-world data" ... This is surprising sin
261.
by n00b101 13y ago
R?
262.
by n00b101 13y ago
Betteridge's law: "Any headline which ends in a question mark can be answered by the word no."
263.
by n00b101 13y ago
> You are not gonna be able optimize it as much as GCC. No one will use. If LLVM were used as a backend for a C11 compiler then optimizations would be for free. Since C is has such a small surface area, and since LLVM is very much design
264.
by n00b101 13y ago
WebRTC is to messaging/socket communication APIs as WebGL is to 3D graphics APIs? So big game changer is that web browsers will implement functionality that native applications have had for decades? I don't understand this obsessi
265.
by n00b101 13y ago
This is how life insurance companies operate on a daily basis. OP had a pretty shitty deal but he should count his lucky stars that he didn't have to interact with actuaries, who would have multiplied the same problems ten-fold.
266.
by n00b101 13y ago
software
267.
by n00b101 13y ago
Next week: linked lists? Linked lists have performance problems on computers. You need contiguous memory arrays for performance, since Direct Memory Access engines, caches, and pipelining all achieve peak performance only when memory is str
268.
by n00b101 13y ago
> If you are doing linear regression with correlated variables. PLS is the correct choice. False. PLS (Partial Least Squares) is an ad-hoc algorithm that was invented by people working in chemometrics. PLS was not based on any rigorous s
269.
by n00b101 13y ago
> This is a good article, but I think it leaves out an important option, which is "do nothing." I don't see how "do nothing" is a good option. If the co-linearity between two variables is high enough, then the li
270.
by n00b101 13y ago
These methods fall under the general categories known as Feature Selection, Feature Extraction and Dimension Reduction: https://en.wikipedia.org/wiki/Dimensionality_reduction

More ›