Some Thoughts

2012 05 26

Words have actual meanings, like eschaton and echelon. I'm a fast reader in that I can get through something and get the general gist of it quickly w/o knowing the precise meanings of a lot of words. Works wonders in marathon long DFW novels, but I realized just now how many levels of awesome I miss out on by being good at extracting meaning from context. For example, the tennis academy kids played a game called Eschaton in Infinite Jest. I spent the entire book thinking eschaton was vaguely similar to echelon and that was good enough. A few minutes ago I realized I didn't really know exactly what echelon meant so I looked it up along with a bunch of other words including eschaton which really means the end of all things, or the final destiny of the world. Nice name for a game about thermonuclear political domination.

2012 05 19

To try to fix my little issue (2012 05 17) with using all the papers I printed out to learn hidden markov models and because I had to sit in a presentation (that really only contained about 10 minutes of material) for 4 hours yesterday I figured I'd give them a shot. Problem is, I took the wikipedia route which sent me to simple markov models which made me realize I didn't understand those enough. I dug in. Here is a free textbook that starts with some very practical applications of markov models in the real world. It's a great intro! Reading that made me realize that I keep forgetting what stochastic means so I thought I'd review today (by typing it out here) to see if I still understand it.

Something is stochastic when it isn't deterministic, meaning it can't be described by a solid equation and always end with the same value. Instead, it will take on a range of values described by some sort of distribution because it is affected by random forces or outside actions.

Think of it as the result of a function.. something like y=mx+b. in a deterministic model, plugging in x and b always gives the same y. in a stochastic model, x and b are not just numbers, they're distributions so the Y is kind of random but it should typically lie within some range. That range can be calculated given the distributions of y and b.

What does that mean for markov models? They're a way to model these sort of non-deterministic problems that vary over time. That is, a stageful model where the current thing moves from state to state with a known probability. I'd try to type out an example, but I couldn't do it justice (just yet) have a look at the first chapter of that book I linked to above if you're interested.

How does this apply to hidden markov models? Well, those are models that you use when you are pretty sure there are hidden, or unknown "states" but you are unable to find the probability distribution of reaching those states.

Reasons why I'm interested: This is a graph based problem which i sort of love. It's rooted deep in probabilities which I am trying to get a handle on because, well, they're useful. And, finally, I have no idea why I'm working on this. I just am. There is this intersection between stats, probability, finance, machine learning, computer modeling and analysis that I'm trying to crack.

2012 05 17

Per the damodaran finance class i'm listening to, the risk free rate basically boils down to the expected rate of inflation in the currency in which you are working. (This is the sound of lightbulbs going off in my head.) Rule of thumb - Use the 10 year zero coupon rate for the risk free rate if the bond is rated AAA. If it is not rated AAA, then take the published rate and subtract off any default risk. You do this because often you will add in default risk at a different point in your analysis. You do not want to double count.

For me, this is kind of huge. I always thought of the risk free rate as the minimum return you could possibly expect with zero risk of default. Which, sure, that's true, but understanding that the market essentially drives that number to track expected inflation of the currency in which you are investing helps me to better understand the fact that different countries have wildly different risk free rates. I never knew why you wouldn't just shift currencies around to get the best deal.. well, i forgot to think about inflation.

update 20120519 - Heh, in lecture 6 he said that the risk free rate is really inflation + macro market growth. I guess that makes sense, too. Still learning

The difference between html's bold and strong tags and italic and emphasis tags eludes me.

Let's talk about hidden markov models. No, really, someone teach me because I ran out of copy paper this week and wound up using all the papers I printed out that I hoped would help me understand them as printer paper. Dammit.

I love John Steinbeck's letter that is on letters of note today:

"If there is a magic in story writing, and I am convinced that there is, no one has ever been able to reduce it to a recipe that can be passed from one person to another. The formula seems to lie solely in the aching urge of the writer to convey something he feels important to the reader. If the writer has that urge, he may sometimes but by no means always find the way to do it."

I know I'm no writer, but everything on this site and armp.it is from me getting this compulsive need (out of nowhere, typically) to put a certain feeling into words and most of the time it doesn't work. Or, rather, I Get It.

I'm putting together a comment service for the app i'm building. I think my implementation is interesting. Here's how i'm doing it. Everything in my system uses a UUID as its ID or primary key. Everything, including each comment. Each comment stores the UUID of the item to which it is attached, the comment text, timestamp, etc. All comments are shoved into this service so that when something says, "I want to display any comments associated with me." All it has to do is pass its UUID to the service which returns a (JSON) list of comments.

Easy peeeezy

The challenge here is how do you go from Comment to the Item. Well, since I don't store the Item's location we fall back to a searching pattern. Every item is indexed anyway, so we just hit the search service with the UUID and get the item back.

This is completely decoupled. It fails if you delete an Item (which, of course, you NEVER delete anything. So, that's not a problem.) Or, if you need to move the comments. Not sure why you would do that, though.

Finally, you can comment on comments. So, your comments can have comments that comment about the comments.

I'm recording guitar parts in my car again while I wait for band practice to start. YOU CAN RECORD DECENT SOUNDING ALBUMS WHILE SITTING IN YOUR CAR. I love the present.

2012 06 05

Real sentences I heard yesterday free of context:

"Queensryche? Must be canadian."

"I just want to know, where should I be going?"

"Good job, girls. Lick that turkey bucket good and clean."

2012 05 14

There are ETF's that track daily s&p returns but are leveraged and inverted. For ex. you can buy an etf that tracks the s&p 3x's inverse. Meaning, if the s&p goes up 10, the etf will go down 30. I didn't read into it much so I'm assuming it is a percentage, and not actual points. it's got to be, right? Anyway: ProShares UltraShort Dow30 NYSEARCA:DXD and ProShares UltraPro SS&P500 NYSEARCA:SPXU seems like a good tool for risk management which leads to:

JP Morgan -> trading complex, highly leveraged, synthetic... hedges? Not sure what to call them yet (still wrapping my head around the syntax) to manage risk can blow up, and everyone makes mistakes, and JP Morgan is trading something like $17T when they're worth something like $14B (that's some leverage.) So, a $2B loss isn't that horrible and it is bound to happen (risk is, by definition, risky.) I won't pretend to know if this kind of thing warrants regulation and as far as I understand of what I've read, they (JP's CIO arm) sort of cornered the market on hedging credit risk which pissed off hedge funds but hedge funds have been fighting regulation so there isn't much they could do but talk to reporters (mid April) which published, which got the attention of other JP arms which ordered CIO to unwind their positions at a costly time and so the pissed off hedge funds are going to sit and wait causing wide bid/ask spreads which will cost JP even more. whoops.

java logging, which framework to choose? After a stack overflow conversation (http://stackoverflow.com/questions/354837/whats-up-with-logging-in-java) I'm guessing I should use slf4j and logback. Done. Today I learned (from code at work) that you can interrogate the api to figure out what appenders/loggers are configured so you can do things to these at runtime. Never gave that any thought till I saw the code and thought, "that's neat." Trying to decide if I want to use JDBC appenders instead of text. didn't have time to dig in to see if 1) they're more performant because if they are I'd like to use them as a deep access log to allow analytics on my method calls and whatnot.

gainseville punk bands and against! me. trying to figure out how much I remember about against! me because they've been trending. I've forgotten more than I know. In fact, I probably only know of against! me because someone gave me a few songs so they're on shuffle and also the admin from cardiology really likes them and told me about a concert she went to last summer. I think that's the extent of my knowledge on pretty much every band until I go to wikipedia and read for 45 seconds and pretend like I know everything about them. yes, that's my roll.

I've given up FB for twitter because I can say stupid things on twitter. To me, Twitter is FB without the cognitive strain of wondering what all these people I only half know from 20 years ago think about my status updates. By design, i never mention anything personal on any social site, until I have to due to well-deserved pressure which is also completely removed from twitter (why would you ever post anything personal on twitter?) but now I have this feeling that 1) i try way too hard to be funny. 2) am i funny at all? 3) people use this thing as a marketing tool to further their careers and to self-promote a well branded image of... themselves. I just talk about meaningless crap and make awful jokes. Is that OK? should I be using social media to put myself in a better light? should I (what are the benefits of) use(ing) social media platforms as a window display of who I think I am as a person? Or, is it that I am, in fact, displaying my true self and what/who I think I am is off mark and I am, in reality, shallow and unable to have an intelligent conversation about the things I think (I think) are important? As a parallel thought, I love to tweet about pop culture and sort of poke fun at it, not because of hate or some self-righteous thing where I sanctimonisouly criticize things people love with a feeling of smug moral superiority making me intolerant of other's opinions and behaviors, but because I genuinely love hearing about what other people (who have feelings) spend their spare time thinking about and seeking out on the internet. Sure, I do not like, "call me maybe" and will never read "50 Shades of Grey" but I love that millions of people do like these things enough to make them a Thing. I love the author of 50 Shades said, in an interview, that she loved Stephanie Meyer's Twilight so much that when she finished reading them she was inspired to write her own books. There is a lot to be said about that. Something about inspiration and believing you can do it to and having the gumption, fortitude and determination to write novels on a goddamned blackberry while commuting to work. I love that. And, for some reason, I express my love for these amazing things that fall completely out of my demographic by mentioning them and poking a little fun, when what I am really saying is, look at me, i know about these Things. This is easier, somehow, than bringing to light things that are important to me.

I constantly google things like "define:self-righteous" because I never know if I'm using words correctly and goddammit, words have actual meanings. you can't just type stuff that sounds right. If I had to guess, i'd say i'm running at about 75% correct usage.

Do an image search for "1957 Ferrari 625 TRC Scaglietti Spider" It is gorgeous.

bonnie (prince) billy (Will Oldham)- a coworker pointed me to this prolific songwriter four days ago. I like his stuff and had never heard of him before. Within 2 days of coming to know about this guy, his name/work has come up in a blog post (one week one band) and as a listentothis on reddit. How does that work? How many times has he come up but I wasn't paying attention? how many great things do I miss or is it that once you hear about something you gravitate toward the sources that likely contain that Thing? Or, is it the Universe (random chance) trying to tell us something. And why is it I randomly thought about the daily coyote (haven't thought about that in at least 2 years) and checked in how she was doing the day she put up a long post about how it is OK to eat meat as long as it (the livestock) is treated really well. Which, fine, ok, that's good. You can't eat wheat, so your diet is already restricted, and you like meat, so ok, justify it, that's great. And Then, 5 days later she announces she used the money from "the daily coyote" book to buy calves which she is now slaughtering for her new company, "Star Brand Beef" which is basically she hugs the cow before she loads it on the slaughterhouse truck where they get sliced up and packaged so she can sell you meat over the internet. There is so much wrong and so much right about that i feel like it balances out to a non-story.

Pandas + rapidQuant = things I would like to spend some time learning but can't 'cause there's just so much. http://www.lambdafoundry.com/rapidquant.html http://pandas.pydata.org/

Also, trying to find time to read Corporate Finance : A Practical Approach by Clayman, Michelle R.; Fridson, Martin S.; Troughton, George H.; Scanlan, Matthew. it is a CFA prep book, but it gets the job done. available electronically via the wustl library but I can't get it when I'ma home.

Manifest file in a jar… mostly useless. Google's guava lib is really good. I can't decide if it and apache-commons libs are complimentary or duplication... I'm using both in the same projects. i probably shouldn't over think this.

Just back from a coworker conversation re: sleep schedules which led to an epic realization. 10 years ago I had a roommate who had a ridiculous 2 hour a night sleep cycle punctuated by a good 20-30 hour sleep binge on the weekend. Sure, she got a ton done and became an MD in the process, but you can't explain that w/o adderal. Never gave that any thought until just now. I don't think she was like that until after she started hanging out with a bunch of young doctors. I was living a real life Jessie Spano I'm so Excited! SBTB episode. I was screech, of course. I'm always Screech...

The only way (I think) to get good at something is to practice it by compulsively working on really hard problems that use the "thing" you want to get good at and seeing them through the end. At least, as many as possible. Anecdote: I got really good at debugging code I had no clue what were doing because of a work-study job at an imaging lab that required me to fix undocumented code that did crazy physics simulations on tagged MRI DICOM data.

She had the kind of face you'd expect to see on a Neutrogena commercial.

2012 05 26

Reading The Real war 1939-1945 and just wanted to keep this in mind:

A little booklet issued to infantry replacements joining the Fifth Army in Italy contained tips to ease the entry of innocents into combat: Don't believe all the horror stories circulating in the outfit you're joining. Don't carry too much stuff. Don't excrete in your foxhole -- if you can't get out, put some dirt on a shovel, go on that, and throw the load out. Keep your rifle clean and ready. Don't tape down the handles of your grenades for fear of their flying off accidentally -- it takes too long to get the tape off. Learn to dig in fast when shelling starts. Watch the ground for evidence of mines and booby traps. On the move, keep contact but don't bunch up. And use common sense in your fight against fear:
Don't be too scared. Everybody is afraid, but you can learn to control your fear. And, as non-coms point out, "you have a good chance of getting through if you don't lose your head. Being too scared is harmful to you. " Remember that a lot of noise you hear is ours, and not dangerous. It may surprise you that on the whole, many more are pulled out for sickness or accident than become battle casualties.

2012 06 13

Today I'm going to present a situation and leave it to you to decide if my reaction was correct.

I like to take a short walk to a coffee kiosk in the hospital where I can get a decent shot of espresso for $1. I toss in a few packets of sugar, let it cool for the walk back to my cube, and it's perfect. The same lady is always there and she's quick, and nice, and doesn't talk much. Perfect.

Today I went around lunch time and she was eating while working. No big deal, although she did say, "what do you want?" with a mouth full of muffin, but, come on, that's funny. She took my money and started my drink. I grabbed a few packets of sugar and waited. She saw said packets and motioned for me to hand them to her. How nice! I passed her the sugar and she ripped them open and poured them in. Then she reached in my espresso with her fingers to retrieve a packet she had dropped. It took her a second, but she got it (how did that not burn?) She let it drip for a second, tossed it in the trash, lidded the cup and passed it to me with a smile. I thanked her graciously for the additional service of adding sugar, walked out of sight and tossed that dirty cup right in the trash.

20120515

I'm recording vocals and guitars for a folk/rock album (the band I'm in is called "The Sorry People") but I only have an sm58 for a mic and an maudio usb interface. I tried some vocals this morning before work and realized the sm58 is not going to cut it. I did some guitar tracks with it and wasn't entirely pleased, but it is good enough. Sadly, not so much for vocals (i need all the help I can get, quality wise) I know I need a good mic, but also thought I needed a good, tube preamp. After doing some research, it looks like as long as you have a good, clean preamp with enough gain, the fact that it has a tube in it doesn't matter as much. I don't know if my cheap maudio interface has enough clean gain for a good condensor mic, but i'm sort of at the point where I don't have a choice. I can borrow a friend's usb interface that may have more/cleaner gain, but I need to source a mic somewhere. I do have access to a blu, but I... I don't think i like how it sounds. I'm going to try to get access to an AKG a friend may be able to let me borrow.... cool story, bro.

I'm working on a list of the bare-minimum requirements for a customer facing web app... all the technologies and features you have to be concerned with before you can open a web-based business. things as simple as an SSL cert to the logic behind signups and payments. There's a lot to think about!

So many shows are about horrible people.

2012 05 16

Let's talk about the kelly criterion. The Kelly Ratio is how much of your bank roll you should bet given a probability of winning, losing, and your edge over the house. This type of strategy can be applied in many different domains, obviously for gaming, sort of for investing but why can't it be applied when estimating other finite resources? Could your "bankroll" represent the "amount of time you can spend" (the finite resource)? Not sure that works because I haven't thought it through, just thinking as i type. I digress. to use the kelly criterion, we have to calculate the player's edge. Say you're playing a game with a dealer where odds are 6/5 and payout is 7/6, or, rather, odds of losing are 1.2 and payout is 1.167. Let's break that down. Say you play the game 11 times. you'll win 5 times and lose 6. Bet $12 each time. You'll wind up with $12x(7/6)=$14 each time you win and $0 when you lose. When you play 11 times, you'll win 5 times giving you 5x$14=$70. you spent a total of $12x11=$132. you lost 6 times which is 6x$12=$72. In other words, in 11 plays you lost $2. edge is loss/total so that's $2/$132 = 0.015 which means the dealer's edge is 1.5%. Your edge is -1.5%. Don't make the bet. Instead, try being on the other side of the bet (become the dealer.)

We're trying to figure out the optimal fraction of our total wealth (bankroll) to bet on each game that will maximize our long term winnings, ignoring volatility but assuring we always have enough $ to bet on the next game (don't allow ruin). The fraction is ((odds received on win)*probability of winning - probability of losing)/(odds received on a win). lets bet on a coin toss where the dealer pays us 100/99 (1.01) on each win. In english, that means for every dollar you bet, they'll give you your dollar back plus one penny if you win. pWin = 0.5, pLose = 0.5 so (1.01*0.5-0.5)/1.01=0.005 = 0.5% That means we should bet 0.5% of our bankroll on each toss. Given those odds and payouts, that makes sense because we get so little for winning. What if we played the same game but the dealer payed us 2/1 for a win? In english, they'd pay you $2 for every dollar you bet. Bet $1, you keep your $1 and they hand you $2 for a total of $3. Common sense says you'd probably want to bet a lot more of your bankroll at that point. Let's do the math. 2*0.5-0.5/2 = 0.25 which means we'd want to bet 25% of our bankroll on each play of the game. Doing so minimizes the chance of ruin and maximizes our total winnings in the long run.

Let's talk about roulette. put $1 on red which pays even money, or 1/1 or 100%. there are 18 red numbers and 20 non red. So, probability of winning is 18/(18+20) = 0.474. plug that into our formula to get 1*0.474-0.526/0.474 = -0.1097... what's that negative number? That means you should take the other side of the bet, or rather, bet AGAINST red using 10.97% of your bankroll. But, there's no betting against red, so you're better off moving along. The casino will always win. (this paragraph is copied from the wikipedia article on the kelly criterion. Go there for more fun information.)

Thinking about the best way to structure my Selenium 2 tests. Need to retrieve the base url from a properties file, of course. probably should have some support inside the application that lets the test avoid certain, untestable-by-selenium features such as email confirmation codes. Maybe the test could log in as admin for a second to grab a copy of the confirm URL so it can be used in subsequent tests. In other words, the app has to support certain features to make automated testing possible. no big deal, but yet another thing I didn't foresee.

Watching shark tank I couldn't off the top of my head how the sharks were valuing a company based on the founder's cash request and the equity offered... I was over thinking it, of course, but here's some things I learned. Typically, a VC wants a 5-10x return on the investment. If I'm a VC and I invest $25k for 25% of a company and expect a 10% return then I want to see a check for $250k when the company exits... That can only happen if the company sells for $1M. If the company sells for $500k then I get a 5x's return so, as a VC, i should still be happy. (rule of thumb, a company's valuation is sometimes 2x's revenue) (Fred Wilson says 1/3 sell for 5-10x's return. 1/3 sell for 1-2x's return and 1/3 fail (0x's return). Since VC's get any leftover cash first, the founders sort of have about a 1/3rd shot of making money, see the paragraph after the next for some discussion.)

But, that doesn't jive with the numbers they throw around on shark tank.. That's because they're not thinking of their return at that time. They're just valuing the current price of the company based on what the founder says. So, if the founder says I want 100k for 25%, the shark tank guys call it a $400k company.

Back to the "founders have a 1/3rd shot to make any money at all" idea. This may or may not be a copy/paste from Fred Wilson's blog. if it is, well, then it is, if it is my words, then great: "A VC is negotiating how much of the upside they'll be in the 1/3 of deals that produce gains. Their deals provide as much downside protection as possible. A VC deal is sort of a loan plus an option where the option part is in the money fully 1/3 of the time and partially about 1/2. loan is repaid fully 2/3 of the time and partially 1/3 of the time."

With that said, I had an idea a few weeks ago. I propose a monte carlo approach to predict project viability and using the results of that calculation as inputs to a binomial pricing model that will let a founder choose among many projects (ideas) to maximize the likelihood of a positive exit. Here's the beginning of an abstract I wrote that day:

"This paper introduces a system for managing uncertain data when predicting market results using a binomial options pricing model to predict exit valuation across multiple projects. This system lets one choose the best projects based on predicted returns using a monte carlo approach to predict project viability."

2012 05 23

I've had no time to think about anything interesting for the past four days. Babysitting, birthdays, Avengers, got rear-ended, did a couple of hard pushes writing boilerplate code. Took a vacation day so I could work my other job. Only got to run once.

2012 05 18

The Lumineers' "Slow It Down" is my favorite song of the day.

2012 09 10

Whoops, I forgot to blog. Read this, especially the part that talks about the Olympus Mons http://www.warrenellis.com/?p=14314

2012 09 12

Ok, so the real reason i haven't been updating this sorry excuse for a blog is i migrated everything to EC2 and I still haven't written an update script to easily publish trickykegstands. Or any of my sites, really.

some links:

2012 05 24

HA! this week has been one disaster after another, but everything still seems to be on course and everyone is pretty much fine. R had a cold, S caught it and started running a fever. We sat with him in the ER from midnight till 4am last night. Now I have the cold. I got rear ended on the day that I had to take a vacation day from one job to work a full day on the other of which I did not get enough work done and should probably take another day or two off. I'll get a new bumper, i guess, eventually. I had to take a few hours off today to recoup from last night's ER visit. As if R didn't already have enough going on taking care of her and S, she lost her wallet so we had to do the whole transfer money out of the ATM account and keep close tabs on the credit card statement, but luckily she found her wallet so that worry is done. Her computer was on the fritz and the parts for-which-she-had-to-drop-everything-to-go-buy were incorrect. So, I wound up fixing the broken part with solder and tape while we ordered a replacement online and had to return the wrong part today on my way in to work (late.) I am falling behind on all 4 of my jobs to the point where all I can do is sit back and laugh. There is so much work I think Keeping Up is an unrealistic expectation. So, I stay the course. That's all I can try to do. I laugh as I type this because even though everything is falling apart, everything and everyone really is fine and there are no real worries. Everything is just a silly mess soaked with vomit and poo right now. Nothing more. This isn't supposed to come off as sympathy seeking or complaining even though I know it does. It is more of a setup for a side-note:

ER waiting areas really are something, aren't they? I work at a hospital so I'm exposed to them (and the places just outside where smokers sit) almost every day. The concern, tears, pacing and phone calls you see happening in these areas make them (the ER waiting areas) feel to me like the front row seat to the last, raw, true glimpse of human beings. Catching a shaky, one ended phone conversation that starts out with, "We lost him" and is punctuated by quick draws on a cigarette by someone sort of hunched over with one arm akimbo probably isn't right but accidentally catching that moment while walking between work areas (really, just cold, grey rooms filled with cubes.) on a generic workday is wow (this is something that someone else can do a better job of describing with real words that mean something but I'm a quitter who settles for wow.)

Last night, after committing various moving violations on the way to the hospital, I dropped S & R off at the door so they could rush inside while I parked the car. I got a good spot then walked with a woman into the ER. She beat me to front desk and asked about so-and-so and the ER nurse said so-and-so was getting a CT and they couldn't say anything more but that there was a whole group of friends there already who knew everything. She pointed to a dozen or so fresh faced HS kids each with that look huddled in chairs, texting and hugging. I eventually found S & R and R said, 'wow, did you see all those kids? I walked in behind 2 parents, i think their son tried to kill himself and those were all his friends.'

wow. yeah, i think my week hasn't been half bad.

old content here

Tricky Keg Stands

A collection of various and sundry things.

Words:

Projects:

Life:

Some Thoughts

2012 05 26

2012 05 19

2012 05 17

2012 06 05

2012 05 14

2012 05 26

2012 06 13

20120515

2012 05 16

2012 05 23

2012 05 18

2012 09 10

2012 09 12

2012 05 24

What is this all about?

Inspirational quote goes here...