Quant Tools- No PhD Required

The first tool you need is a little imagination and endless curiousity.
The second tool you need is diligence.
The third tool you need is basic spreadsheet skills.
The fourth tool you need is good music, exercise, and inspiration.

the need for imagination and curiousity was pretty much covered in the previous post. but a couple of concrete examples might be added... like, have you ever wondered if there's any correlation between how you feel on a given day and the variance in the barometric pressure over the last few days? or why the birds sing louder at dawn some days than others? or why disasters seem to occur in bunches? (maybe there's something to the old saw about bad things coming in three's, after all, huh?). if you've ever wondered about weird stuff like this, you have the basic mentality of a quant (and God help you!).

let's go further. have you ever wondered if stock price is somehow related to some twisted mathematical combination of factors like volume, OS (outstanding shares), and price differences over the past five days or something like that but just didn't have a way to compute it? if so, then again you are a confirmed quant. and you're in deep trouble because it's absolutely addictive: sooner or later you'll begin quanting your kids' behavior patterns, your spouse's spending habits, your girlfirend's/boyfriend's likelihood of calling you at a certain time, and doing a hundred other annoying things.

so you see that, in a way, quanting is definitely obsessive and borderline psychotic behavior. don't worry, though- there are checks and balances, like that first icy glare from your loved one that tells you you've gone 'way too far into wonderland and it's time to come back. or, when you look up all bleary eyed and grimy and realize you've been hunting an algo for the same stock for two days running and figure out that you could've spent the time actually finding working algos for two dozen other stocks. it's a long road but you'll get there.

how do i know all this? let's just say i'd rather not say...

diligence is not necessarily about obsession, though sometimes they're the same thing. the kind of diligence you need in quant is to set yourself some quanting rules and be diligent in observing them. for example, you might tell yourself...

• I'll only quant on weekends, for 4 hours each day between the hours of 6pm and midnight.
  
• I will not continue unsuccessful attempts to quant any single stock for more than 30 minutes.
• I will add at least one stock with a working algo per day.
• When quant trading, I will follow my entry/exit plan- no excuses.
• I'll backtest my working algos every two weeks to be sure they still work (algos do change).
• I will not give out my picks except to a couple of trusted associates (reasons covered earlier).
• I will try at least two new algos per work session, and not get stuck in a rut.

it's all about style- pick a style that works for you and stick to it. that's how you make steady progress and stready profits. to have a ripe trade every day, you need about two dozen quanted stocks with working algos. updating with current data takes about 2 minutes per stock per day, or the same amount of time if doing only weekly updates (in which case you'll miss a few ripe plays).

i want to skip now to the 4th tool- good music, exercise, and inspiration. it's far more important than most people realize. first off, rent a copy of The Hustler and watch it (Jackie Gleason with Paul Newman). it leaves a lasting impression about the importance of staying fresh. Minnesota Fats beat the Hustler in that billiards tournament because Fats kept himself fresh and the Hustler didn't. simple as that.

take a brisk walk before you start quanting. eat lightly (or walk briskly afterward). watch scientific pieces on tv (like NOVA, etc.). these will inspire you. read scientific stuff, even stuff you think is over your head. i read Nature and Scientific American, even though i often don't understand more than 10% of what the heck they're talking about. in fact, one of my best algorithms came from that- just one sentence that i did understand. it turned on a light inside my head but the article was about biology, for crying out loud!

music. here's my personal drill: Janis Joplin, Michael McDonald (Heard It Through The Grapevine), Aretha Franklin, Tina Turner, some Crosby, Stills, Nash, and Young, and some smooth jazz. i load it up, crank the volume, and pull the trigger, making instant enemies of my neighbors. a half hour of that then i turn it down low (or off) and coast on the adrenaline rush while i do my quant thing. i say pick music that gets your blood up, but...

aaaagggghhh- as a devout child of the 60's i hate to admit this... but scientific studies have shown that classical music is better for organizing the mind. ugh! i hated typing that, but as a scientist and physician i have to accept it. now my keyboard is going to bite me.

(continued in next blog...)