Quant

« first day (8 days earlier) ← previous day next day → last day (3033 days later) »

Dirk Eddelbuettel

01:43

@chrisaycock Here you go. The noob factor has certainly increased. Wonder if the VBA+Excel guys will be next after the IB crowd...

3 hours later…

Shane

04:23

@DirkEddelbuettel VBA/Excel? Isn't that your specialty?

Dirk Eddelbuettel

@Shane How did you know? With all my Ruby/Javascript covers...

I can't get my head around Quora either. But then again I don't why I help Spolsky build something he will one day sell for a handy sum...

2 hours later…

chrisaycock

06:18

@DirkEddelbuettel Well, is a web developer helping Google increase its share price? I guess what we're aiming to do is create a discussion for actual quants. We've had enough false starts with Wilmott, Nuclear Phynance, EliteTrader, etc. The only thing that comes close is QuantNet, though that's aimed more at MFE students than working professionals. This is finally a chance at our thing.

Quora is interesting since I can ping someone to answer a question. I got one of the guys from Marketcetera to answer my question about his business model.

I also managed to answer a question about Y Combinator that a YC partner up-voted, which I thought was cool.

But I'm on several sites anyway, including Hacker New and Quantly (from my old boss), so I don't feel like I'm exclusive to just Spolsky.

15 hours later…

Dirk Eddelbuettel

21:33

I think between "TheBridge", "quant_dev" and "user40" we have a number of dodgy questions and answers. Or am I too picky?

Shane

22:06

@DirkEddelbuettel No, you're right. Vote to close and comment on them.

chrisaycock

I wish there were a throttle for users who post three or four questions in a row.

Joshua Ulrich

I like whoever asked Shane to leave the board alone. :)

chrisaycock

@JoshuaUlrich Which one was that?

Joshua Ulrich

Q: What are the tools fund managers use most?

By tools I mean both software and statistical/mathematical tools to help them do their job.

chrisaycock

Ah, right.

Dirk Eddelbuettel

22:13

I'll follow up in due course with 'Why is the sky blue'.

chrisaycock

While Quant.SE is still in its infancy, Shane has been one of the driving forces to define this site. I think we should cut him some slack.

@DirkEddelbuettel Oh, I'll post, "What's a profitable model I can use?"

Joshua Ulrich

I was being sarcastic. I agree with Shane's vision for the site.

chrisaycock

@JoshuaUlrich Ah, ok. That user (ChloeRadshaw) was the one who posted the question about how banks make money on mortgages. I wish there were a "block all posts from this user" button.

Joshua Ulrich

@chrisaycock Yes and all of their questions had been closed, which is why I find it funny that they ask Shane to "leave the board alone".

Shane

Wow. Didn't see that one.

Just posted a comment response.

Joshua Ulrich

22:20

Drama, I love it!

Shane

It's interesting, I'm developing a slightly bad reputation here. Someone maligned me on meta before. Basically, I don't want to allow bad questions. Sue me.

chrisaycock

Did the other forums (NP, Wilmott) ever try to get rid of the beginners? How can we avoid turning into those places?

Shane

By closing bad questions and telling people these questions aren't on-topic. That's why CrossValidated works.

I don't think that NP/Wilmott do a good job there; there isn't really a mechanism for it.

Joshua Ulrich

@Shane Did you have similar growing pains with CV?

chrisaycock

How was CrossValidated in its infancy? Did it have a swath of "what's the difference between median and mean?" questions?

Shane

22:23

For sure. Lots of stupid/general questions.

This will calm down here as we keep plugging away.

The key is to get a few moderators on the site who have a similar vision.

Dirk Eddelbuettel

int rand(void) { return 3; } // why doesn't this work?

chrisaycock

A little XKCD humour?

Dirk Eddelbuettel

Well spotted sir.

Joshua Ulrich

@DirkEddelbuettel Because only even numbers are random...

Shane

That was impressive.

Dirk Eddelbuettel

22:24

I do recall recall an econometrics textbook defining an estimator in the intro as 'always return 42'

Shane

I'll give it the first ceremonious star. Like a boss.

Dirk Eddelbuettel

Which is awesome. Totally consistent estimator. Unfortinately a tad biased.

Shane

Ok; hitting the subway. Talk to you chaps later.

chrisaycock

Dirk and Shane, did you guys know each other before from CV and SO? You're the top R users, from what I recall.

Shane

I think that we were introduced through SO, actually.

Dirk Eddelbuettel

22:26

I will not admit it in public, but yes, I do (virtually) know this man.

chrisaycock

Well put, Dirk.

Shane

I formed some good connections through CV as well. I'm hoping the same will happen here.

Only time will tell, but we're off to a reasonable start.

chrisaycock

Cool.

I'm hoping the same with Quantly. My old boss started that as a "Hacker News for quants" (mostly quant traders in practice). That hasn't been hit with too much beginner cruft, fortunately.

Joshua Ulrich

I'll try to think of some answerable, non-MBS questions. I'd like to contribute more.

chrisaycock

I look forward to that.

Joshua Ulrich

22:31

Don't set your hopes for me too high; you'll likely be disappointed. ;-)

chrisaycock

Well, the one question I asked was all set to get a Tumbleweed until I offered a bounty.

Joshua Ulrich

ouch, that's unfortunate

Dirk Eddelbuettel

@chrisaycock Because it was friggin hard !

chrisaycock

I didn't expect a full proof (especially since I didn't give a rigorous definition of what a meta model is), but I thought at least somebody could give some intuition.

I don't know your backgrounds (I'm assuming statistics), but my academic field was founded because of a question about decidability.

Entscheidungsproblem

In mathematics, the ' (, German for 'decision problem') is a challenge posed by David Hilbert in 1928. The asks for an algorithm that will take as input a description of a formal language and a mathematical statement in the language and produce as output either "True" or "False" according to whether the statement is true or false. The algorithm need not justify its answer, nor provide a proof, so long as it is always correct. Such an algorithm would be able to decide, for example, whether statements such as Goldbach's conjecture or the Riemann hypothesis are true, even though no proof or d...

Joshua Ulrich

My background is in economics.

Dirk Eddelbuettel

22:40

We are pretenders. Oh, wait, Shane, like a boss, is a real engineer.

chrisaycock

I see.

Joshua Ulrich

Dirk's a real pretender. I only pretend to pretend.

Dirk Eddelbuettel

If you're on LinkedIn, the sort-of minimal bio is there for me too.

chrisaycock

I had to get off LinkedIn since I was getting hounded by head hunters.

I was like the hot girl at the bar, apparently.

Dirk Eddelbuettel

That's the only reason we're talking to you here too.

chrisaycock

22:47

I knew it!

Joshua Ulrich

23:08

@chrisaycock can I buy you a beer? ;-)

« first day (8 days earlier) ← previous day next day → last day (3033 days later) »

all rooms

Transcript for

Feb8

Feb '119

Feb10

Discussion for quant.stackexchange.com

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC