Computer programmers/traders

[ThingyMajiggy]

Just a heads up guys, we've had a ripper hipster beard pattern play out on the 15m CL chart, get on it!

 

[DeepState]

Are the hipster beard lengths weighted by social ranking of the hipster? How do you normalize for alternate measurement techniques by said barber shops?

Data transparency please!

Are you taking the piss?

Yet again KAO is claiming my work as hers. If it weren't for her skills, I would be upset at her.

Anyhow, this is my field. So let me explain:

The link between Arabica beans and beard length is absolutely obvious. But in case it isn't the idea is that there is a conclusive link between testosterone and caffeine as per Beaven et al (2008) which researched the link from the perspective of sport and use of drugs to enhance performance. Beard growth is associated with testosterone. Hence, more beards are associated with more caffeine...Arabica beans go up.

Beards are relatively slow to grow and show lags...or momentum. Hence, you can spot trends and extrapolate with surprisingly high accuracy. It also helps that coffee drinking habits show considerable state rigidity. If you start drinking 2 cups per day, you tend to stay at two cups. If you drink espresso today, it tends to be espresso tomorrow.

More beard growth...a leading indicator of arabica demand. You only need to have an edge in this, not the supply side, to make a ton.

That stuff is freakin' easy. A child could do it, so long as it has a 180 IQ.

The really hard stuff is standardising beard length and growth rates for different barbers. Beard growth and propensity to grow it varies by age, education, income, gender (more on this later) and ethnicity. So, in order to know if there is an acceleration in beard growth rate, you need to allow for this by the post code. The multifactor regression also has to be HAC consistent because some barbers are crap at measurement.

---

That stuff is easy. We were doing it in 1995, but the signal really took off in the last few years for obvious reasons.

Kao and I, mostly me, made the following advances for the basic application. This is the sweet stuff.

You need to allow for daily seasonality in the barbershop measures. Most people go on a particular day of the week each time they have a trim. That cohort exhibits clear seasonality. The short beard guys want a trim before going for parties on a weekend. Hence there is a bias to short beards on Fridays. Long beard guys could care less. They are all over the place. Making this seasonal adjustment, the State-Kao seasonal oscillator, slices the Sortino ratio in half...not just trimming it...haha.

The other breakthrough was that, as the general age of the population grows and ages, women also grow beards. The data does not distinguish on gender as no-one allowed for it when it was being collected for Qandl. This is the stuff of opportunity. So, it turns out, that the census can be used to further partition the data to allow for this effect. Amongst those who know this stuff, it is called the State-Kao demographic support vector machine. Snip...off comes the Sortino again.

---

So let me be clear that this particular signal is rather more mine that KAO's. Nevertheless, KAO has done some decent work in the past which I won't do her the disservice of revealing to a public forum without thought or care for alpha crowding. I would add, though, that KAO really is amongst the more hard core amongst the truly wild guys/gals in algo land.

If you have something to offer, we welcome collaboration for mutual profit and enjoyment. Where's that umbrella?

 

[traderxxx]

is there a way for members to contact other members directly?????

 

[ThingyMajiggy]

is there a way for members to contact other members directly?????

Click their profile picture, then "Start a conversation".

 

[traderxxx]

how does that work, does it go to their email
or a conversation on the forum, in which case they have to
be watching.

 

[ThingyMajiggy]

how does that work, does it go to their email
or a conversation on the forum, in which case they have to
be watching.

It's a private message on the forum, they'll get a notification(top right, little red box when you have a new notification) and they can go read it in their inbox. They'll probably get an email notifying them about it as well, but it's all based on the forum. Only you two will be able to see it.

 

[traderxxx]

ok thanks ThingyMajiggy,

 

[barney]

Anyhow, this is my field. So let me explain:

Beard growth is associated with testosterone. Hence, more beards are associated with more caffeine...?

LOL, very funny 'DS' ..... I suspect that post actually made a lot more sense than I realised

 

[traderxxx]

thanks for the help ThingyMajiggy.

 

[galumay]

If you have something to offer, we welcome collaboration for mutual profit and enjoyment. Where's that umbrella?

I think you and the pretty Ms KAO may have missed some important signals in the Arabica market, I see no mention of brakeless, single speed bicycles, when you talk about momentum these are critical to proper analysis, their cyclical nature must be cconsidered too.

The problem with relying on beards is they are not restricted to pretentious, but largely skill less baristas (aka hipsters) who wouldn't know anything about Rao's ratio or extraction effeciencies let alone TDS levels. You have false positives on hippies, bikies, old Texans and NZ mussels.

I suspect that combining the mad skillz of Ms RAO, DS and me with my experience as a professional coffee roaster who can spot a hipster on a fixie, at night, with my eyes closed - would be a prodigious trading partnership.

 

[barney]

You have false positives on hippies, bikies, old Texans and NZ mussels.

Whatever kind of Bourbon you guys are on .... send me a bottle will ya!

At first I was skeptical ... but its true .... NZ mussels do have beards!

 

[jenisa]

The title brought me here. Having a combination of the two would be an absolute dream. Hahaha!