Multicollinearity revisited
So as none of you probably remember, I did a blog in April (March?) on the perils of dealing with data that had multicollinearity issues. I used a lot of Venn diagrams and a lot of exclamation points.
Today I shall rehash my explanation using the magical wonders that are vectors instead of Venn diagrams. Why?
1. Because vectors are more demonstrably appropriate to use, particularly for multiple regression,
2. I just learned how to do it, and
3. BECAUSE STATISTICS RULE!
So we’re going to do as before and use the same dataset, found here, of the 2010 Olympic male figure skating judging. And I’m going to be lazy and just copy/paste what I’d written before for the explanation: the dataset contains vectors of the skaters’ names, the country they’re from, the total Technical score (which is made up of the scores the skaters earned on jumps), the total Component score, and the five subscales of the Component score (Skating Skills, Transitions, Performance/Execution, Choreography, and Interpretation).
And now we’ll proceed in a bit more organized fashion than last time, because I’m not freaking out as much and am instead hoping my plots look readable.
So, let’s start at the beginning.
Regression (multiple regression in this case) is taking a set of variables and using them to predict the behavior of another variable outside the sample in which the variables were gathered. For example, let’s say I collected data on 30 people—their age, education level, and yearly earnings. What if I wanted to examine the effects of two of these variables (let’s say the first two) on the third variable (yearly earnings) That is, what weighted combination of the variables age and education level best predict an individual’s yearly earnings?
Let’s call the two variables we’re using as predictors X1 and X2 (age and education level, respectively). These are, appropriately named, predictor variables. Let’s call the variable we’re predicting (yearly earnings) Y, or the criterion variable.
Now we can see a more geometric interpretation of regression.
Suppose the predictor variables (represented as vectors) span a space called the “predictor space.” The criterion variable (also a vector) does not sit in the hyper plane spanned by the predictor variables, but instead it exists at an angle to the space (I tried to represent that in the drawing).
How do we predict Y when it’s not even in the hyper plane? Easy. We orthogonally project it into the space, producing a vector Y ̂ that lies alongside the two predictor variable vectors. The projection is orthogonal because we want to make the angle between Y and Y ̂ the smallest in order to minimize errors.
So here is the plane containing X1, X2, and Y ̂, the projection of Y into the predictor space. From here, we can further decompose Y ̂ into portions of the X vectors’ lengths. The b1 and b2 values let you know the relative “weight” of impact that X1 and X2 have on the criterion variable. Let’s say that b1 = .285. That means that for every unit change in X1, it is predicted that Y would change by .285 units. The longer the length of the b, the more influence its corresponding predictor variable has on the criterion variable.
So what is multicollinearity, anyway?
Multicollinearity is a big problem in regression, as my vehement Venn diagrams showed last time. Multicollinearity is essentially linear dependence of one form or another, which is something that can easily be explained using vectors.
Exact linear dependence occurs when one predictor vector is a multiple of another, or if a predictor vector is formed out of a linear combination of several other predictor vectors. This isn’t necessarily too bad; the multiple or linearly combined vector doesn’t add much to the analysis, and you can still orthogonally project Y into the predictor space.
Near linear dependence, on the other hand, is like statistical Armageddon. This is when you’ve got two predictor vectors that are very close to one another in the predictor space (highly correlated). This is easiest to see in a two-predictor scenario.
As you can see, the vectors form an “unstable” plane, as they are both highly correlated and there are no other variables to help “balance” things out. Which is bad, come projection time. In order to find b1, I have to be able to “draw” it parallel to the other predictor vector, which, as you can see, is pretty difficult to do. I have to the same thing to find b2. It gets even worse if you, for example, were to have a change in the Y variable. Even the slightest change would strongly influence the b values, since when you change Y you obviously chance its projection Y ̂ too, which forces you to meticulously re-draw the parallel b’s on the plane.
SKATING DATA TIME!
Technically this’ll be an exact linear dependence example (NOT the stats Armageddon of near linear dependence), but what’re you gonna do?
So bad things happen when predictor variables are highly correlated, correct? Here’s the correlation matrix for the five subscales:
I don’t care who you are, correlations above .90 are high. Look at the correlation between Skating Skills and Choreography, holy freaking crap.
So let’s see what happens in regression land when we screw with such highly correlated predictor variables.
If you’re unfamiliar with R and/or regression, I’m predicting the total Composite score from the five subscales. The numbers under the “Estimate” column are the b’s for the intercept and each subscale (SS, T, P, C, and I). But the most interesting part of this regression is under the “Pr(>|t|)” column. These are the p-values which essentially tell you whether or not a predictor significantly* accounts for some proportion of variation in the criterion variable. The generally accepted cutoff point is any value p < .05 (with anything less than .05 meaning that yes, the predictor variable accounts for a significant proportion of variance in the criterion).
As you can see by the values in that last column, not a single one of the predictor variables is considered to be significant. Which is odd, when you think about it initially—after all, we’re predicting the total Composite score, which is composed of these individual predictor variables—why wouldn’t any of them be significant in terms of the amount of variance they account for in the Composite? Well, because the Composite score is the five subscales added together, it’s a direct linear combination of the predictor variables. Because all of the predictor variables appear to account for an equal amount of variance—and because the variance in the Composite score involves a lot of “overlapping” variance from each of the predictors, none of them are deemed statistically significant.
Cool, huh?
*Don’t even get me started on this—this significance is “statistical significance” rather than practical significance, and if you’re interested as to why .05 is traditionally used as the cutoff, I suggest you read this.
Today’s song: Top of the World by The Carpenters
Was Ockham’s razor a Gillette?
This is fun.
Trait snapshot: depressed, introverted, neat, needs things to be extremely clean, observer, perfectionist, not self revealing, does not make friends easily, suspicious, irritable, hates large parties, follows the rules, worrying, does not like to stand out, fragile, phobic, submissive, dislikes leadership, cautious, takes precautions, focuses on hidden motives, good at saving money, solitary, familiar with the dark side of life, hard working, emotionally sensitive, prudent, altruistic.
Today’s song: Sick Muse by Metric
Netflix, eh?
HOLY CRAP Netflix.ca has happened!
Awesome.
Also, here’s a quiz that told me what type of cheese I am. I’ve posted this before, but it was long ago and my cheesy ways have changed (I used to be brie).
Today’s song: Falling Inside the Black by Skillet
I downloaded Steam and it downloaded my soul
So for all the PC gamers out there, I totally recommend downloading Steam. Sean was trying to get me to do so for the longest time and I never did, but for whatever reason, today I decided I *needed* to play around with Garry’s Mod, so I downloaded Steam and verified my old Half-Life CDs.
It’s pretty rad.
ALSO I can now play Deathmatch Classic, which is like my favorite thing ever. I also totally own at it, too.
Today’s song: End Love by OK Go
I AM BLOGGER, HEAR ME POST
Claudia’s Awesome Salad
Hey ladies and gents. Today I shall present you with a recipe for salad. Because it’s a freaking awesome salad.
Ingredients you shall need:
- Broccoli (1 ounce – I don’t know how many little florets this is; I’m picky about the way I cut my broccoli. Guesstimate or use a food scale)
- Carrots (1 ounce – approximately three baby carrots)
- Radishes (two medium-sized ones)
- Cauliflower (2 ounces – approximately two large florets)
- Shredded parmesan cheese (3 tablespoons, or to taste; I like cheese)
- Lettuce (100 grams – I use iceberg lettuce)
- Croutons (20 grams – about a small handful)
- Dressing (2 tablespoons – I use Kraft Calorie Wise Caesar)
Use a cheese grater to grate the carrots, radishes, and cauliflower (I just run the whole cauliflower florets over the grater, it seems to work best that way) and combine with the broccoli, chopped any way you prefer. Add the parmesan and stir. It should look something like this:
Pick apart lettuce to manageable sizes and add to the mixture. Toss to mix everything up. Add the dressing and toss again, then finally add the croutons. Voila!
Note: this makes a lot of salad, but depending on the type of dressing/croutons/cheese you use, it can still be pretty healthy. My version runs about 250 calories (10 grams of fat, 4 grams of fiber, almost 11 grams of protein), which is pretty good for the volume. I eat this and Mini Wheats for dinner ‘cause I’m weird.
Woo!
Today’s song: The Fame by Lady Gaga
My Internal Dialogue is in All Caps
So I had this super awesome statistics-related blog all planned out to post tonight, but a friend in Multivariate Analysis had to borrow my notes, so I figured I should wait until I get them back to make sure I don’t make any mistakes regarding the example I’m going to use.
So tomorrow. I promise. I also keep forgetting to update this more often than whenever the hell I feel like it; I apologize, I’m still conditioned from MySpace.
In the meantime, here’s a color sample of all the clothes in my closet.
Scary, huh?
Today’s song: Sometime Around Midnight by The Airborne Toxic Event
This just in: Cap’n Crunch promoted to Adm’ral
Cereal mascots have always fascinated me. Many of them are poor, deprived souls who just want to try the product they endorse (but are prevented by kids, circumstance, sugary villains, etc.). Others most certainly have mental issues brought on by the cereal they consume (a certain bird who really enjoys cocoa comes to mind).
And then there’s Cap’n Crunch.
First off, his full name is Captain Horatio Magellan Crunch, which is about as awesome as a name can get for a guy with no neck, a freakish cereal fetish, and who has been stuck at the rank of “Cap’n” for god knows how long.
Don’t get me started on the commercials. Says Wikipedia, “In modern TV ads, Cap’n Crunch is often seen riding his ship through a wall as the whistle blares.”
Sooo…Kool-Aid Man of the sea?
And: “He often comes in the middle of a predicament and uses his cereal to solve the problem at hand by ‘Crunch-a-tizing’ it.”
What I wouldn’t give to have that power.
Lab manager: Oh crap, SPSS is being a bitch (again) and we can’t get these analyses done!
Lab member: Our assignment is due in three hours! What are we to do?!
Me: *breaks down Kenny wall in a ship*
Lab manager: What the hell…?
Me: CRRRRRRUNCHATIZE!
Me: Join me for some high-sea, high-fructose fun!
Lab member: How would that possibly solve our problem?
Me: Cap’n’s orders!
*dumps cereal everywhere*
Lab manager: I doubt you’re a licensed mariner.
Lab member: Where did you get all this cereal?
Me: Set sail for dairy goodness!
*unleashes gallons of milk*
Lab manager: You destroyed my laptop!
Me: OOPS! All Berries!
Etc.
Yeah. I think he’s cool.
Today’s song: The Scientist by Coldplay
Stop Waiting for Godot! Take ACTION!
Because someone finally tagged me in one of these on Facebook, but I don’t write notes, so I put it here instead.
Plus I’m bored.
Official Rules: Once you’ve been tagged, you are supposed to write a note with 16 random things, facts, habits, or goals about you.
- Some people think I’m over exaggerating with the whole “I love Leibniz” thing. I’m not. I’d take that man in a second if he were alive today.
- I really, really, REALLY don’t like people touching my bangs.
- Plugging my iPod into the car stereo, cranking it, and driving around for hours = major stress relief.
- I try to give at least one compliment a day—if not verbally, than in my mind.
- Back in 1st/2nd grade, I’d construct life-size paper people out of construction paper. They had bones (popsicle sticks), blinking eyes, “working” digestive systems, and could wear my clothes. I also made elaborate little paper laptops during recess time.
- It really bothers me when people judge others based on their musical tastes.
- I freaking love McDonald’s French fries. I don’t care that they’re pretty much poison, I think they’re absolutely fantastic. Especially when they’ve been sitting under the heat lamp for a little too long.
- I haven’t worn a pair of shorts since 2nd grade.
- I haven’t worn a pair of jeans since 8th grade.
- The majority of my pet peeves are concentrated around the actions of self-centered people. You know, the people who ALWAYS direct the conversation back to themselves, the people who decide not to acknowledge someone waiting for an open locker in the changing room and instead stand in front of their locker once they’re done with their workout and text for ten minutes, the people who walk down the middle of a narrow walkway without realizing that perhaps someone would like to pass them…the list goes on.
- I can’t remember what my life was like back before I’d discovered my passions for statistics and philosophy. I must have been very miserable.
- My first word was “tick-tock.”
- I was a fantastic runner/jumper in elementary school (I wanted to go to the Olympics). Now running hurts just about every square inch of my body for whatever reason.
- There are some days when I have this incredible compassion towards the whole of humanity. There are other days where I just want to stab everyone.
- I would love to be a chef, but unless my olfactory bulb suddenly decides to work for the first time ever, I don’t see that ever successfully happening.
- I almost decided against going to college.
WOO!
Today’s song: The Tip of the Iceberg by Owl City
TWSB: You are my sunshine, my only sunshine…you may determine how I decay…
More news pertaining to our star for this week’s science blog: apparently scientists from Perdue and Stanford have found that the decay of radioactive isotopes fluctuates with the rotation of the sun’s core.
The fluctuations are small (and most likely won’t radically alter any anthropological findings), but they may lend a hand in predicting future solar flares as well as have an impact on medical radiation treatments. The scientists have been collecting data for nearly four years and have determined (at least in the cases of silicon-32 and chlorine-36) that decay rates follow a 33-day pattern.
So how the hell can the sun affect decay rates? The scientists believe it’s due to solar neutrinos, near weightless particles produced by nuclear reactions in the sun’s core. However, these neutrinos have never been known to actually interact with anything before, so one of the scientists summed things up in the rather humorous sentence, “So, what we’re suggesting is that something that can’t interact with anything is changing something that can’t be changed.”
Woo!
Today’s song: Robot Rock by Daft Punk
AAAAHHJESUSJESUSHOLYFREAKINGCRAP
I…I have no words. These topics are…me. Almost every single one of the articles/chapters on this page is something I’d kill to study.
This wonderful, glorious human being is a professor at the University of Wisconsin-Madison. His research interests include general philosophy of science, philosophy of physics, and philosophy of statistics. Therefore, I must apply to the University of Wisconsin-Madison.
MY DESTINY LIES IN THE MIDWEST!
I’m off to go dance around now. This made me more excited than anything probably should.
Today’s song: U + Ur Hand by P!nk
10 Reasons Why Everyone Should Love Quake
Quake rules and here’s why:
10. When you’re totally out of ammo, you still get an axe
Almost as good as the crowbar in Half-Life (though nothing will ever beat that), when you’ve exhausted all ammo (Shamblers, anyone?) you’re left with a little bloody axe that you swing like a dork. It’s freaking great.
9. Scrags
I love Scrags, and I’m not really sure why. I remember we had to make soda bottle water rockets in 5th grade, and I managed to decorate mine to look like a Scrag (obsessive much?). There’s a level that’s almost entirely Scrags; I like to go into God mode and just play with them. Yeah, I’m that cool.
8. You kill the final boss by waiting until a purple spiked ball floats through her
To my little first grade mind when I first played through Quake, this was so freaking amazing. You had to time it so that you went through a teleport gate as the little spiky goes through her body. While you’re surrounded by Shamblers. And lava. Yay.
7. Story line? Pfft.
That was the good thing about mindless FPS games back in the early nineties—they were mindless. I like shooters, especially when there’s no other point than to see how good you can be with strafing while shooting. When (actual) story lines are developed, it loses some of the genre’s charm.
6. Cheating is super fun
No clip activated in water + God mode = JESUS FLIGHT! I always used to fly up and out of the map, or into the weird little ceiling textures above some of the upper levels. It was great. I really need to play more Quake.
5. The Nailgun(s)
DUDE I LOVE THESE. I got my gamer name (Nailpit) partially due to these guns. There’s nothing more satisfying than firing a crapton of nails at stuff. Except maybe being Gordon Freeman.
4. Shamblers
These things scared me when I was little. I really, really didn’t like them. They shoot lightning bolts from their hands and make this awful guttural growl. Plus they’re one of the hardest enemies to kill.
3. Quake is perfect for speedruns
Quake Done Quick and The Rabbit Run are two very nice speedruns through all the levels of Quake, proving that with games like these, run-throughs can be done at a ridiculously fast pace. Pretending to be good at doing so is pretty fun, too.
2. Quake begot Half-Life
And we all know how awesome Half-Life is.
1. The fact that you can play it effectively with just arrow keys, a spacebar, and the control button
This is my favorite component of Quake. It has a y-axis, but you don’t really need to use it. “Aiming” is essentially accomplished by pointing your weapon in the general direction of the enemy and firing; there’s never any real need to look up or down, unless you’re paranoid about platforms or possible enemies on floors above.
I guess I like it ‘cause you don’t need a mouse to play it, just a bit of finger dexterity on the keyboard. That appealed very strongly to my first grade mind, and now that the majority of games I play are on the Xbox 360 or are PC games that require the use of y-axis looking, I really appreciate the simplicity of “up, down, right, left, spacebar jump, control key shoot” gameplay.
Woot.
Today’s song: Caramelldansen by Caramell
How many grad students does it take to configure an office space?
Answer: Three. One to try and decide in what arrangement the desks should be, one to move said desks, and one to realize, after all possible combinations of desk arrangement have been attempted, that they’ve decided on keeping the desks in the order they were originally.
Yeah.
Related news: I’m moving to the Botany Annex, ‘cause the Social/Personality area needs more space. It’s okay, though, ‘cause in the Annex the windows actually open.
Woo.
Today’s song: Put Your Hands Up for Detroit (Radio Edit) by Fedde Le Grande
It’s a post! It’s a blog! It’s a survey!!
10 things you love
1. Statistics.
2. Philosophy.
3. Leibniz.
4. R.
5. Writing.
6. Color.
7. Analyzing stuff.
8. Puns.
9. Antarctica.
10. The internet.
9 talents
1. Obsessing.
2. I’m very good at packing a lot of stuff into very small spaces. I should be a professional mover.
3. Bending R to my will.
4. Rocking it at Quake.
5. Writing (or so I’d like to think).
6. I’m pretty damn good at principal components analysis.
7. Starting Flash projects and never, ever completing them.
8. Dressing like I stood in the blast zone of an exploding Crayola factory.
9. Can blogging count?
8 favorite people
1. Sean.
2. Aaron.
3. Matt.
4. Maggie.
5. Rebeca.
6. Nick.
7. Jacob.
8. Aneel.
7 goals
1. Accrue as much knowledge as possible.
2. Get a PhD.
3. Have a career in which it’s part of my job to screw around with R.
4. Do something with Prime, aside from let it sit untouched on a flash drive.
5. Visit Antarctica.
6. Finish my book list that I’ve been working on since 7th grade. I have a lot more time to read now that I’m not taking 8 classes a semester.
7. Make a metric ton of money and then do something good with it.
6 things you think about a lot
1. Determinism.
2. Where I can find data to analyze.
3. Trends and patterns (in pretty much everything).
4. Philosophy of science-related stuff.
5. What the next thing I can draw/write/create should be.
6. What to blog about.
5 favorite songs
1. Passion Pit’s Sleepyhead (and all its beautiful, beautiful variants).
2. Dan Black’s Symphonies.
3. Morten Lauridsen’s O Magnum Mysterium.
4. Leisure Alaska’s Hey There Mr.
5. Battles’ Atlas.
4 worries
1. My MA thesis. It scares me.
2. The next 4-5 years of my life. They scare me.
3. Mental health “issues.”
4. What the next day will bring.
3 things you believe in
1. A deterministic universe.
2. Logic.
3. Hard work.
2 best experiences of your life
1. Finishing my first degree in 2 ½ years with a 4.0. I worked so damn hard for that.
2. A certain night that won’t be mentioned by name, but you all probably know what it is…
1 thing you want right now
1. Reassurance.
Today’s song: Breakeven (Falling to Pieces) by The Script
Mmm, fresh data!
Hey ladies and gents. NEW BLOG LAYOUT! Do you like it? Please say yes.
Anyway.
So this is some data I collected in my junior year of high school. I asked 100 high schoolers a series of questions out of Keirsey’s Please Understand Me, a book about the 16 temperaments (you know, like the ISFPs or the ENTJs, etc.). When I “analyzed” this for my psych class back then, I didn’t really know any stats at all aside from “I can graph this stuff in Excel!” (which doesn’t even count), so I decided to explore it a little more. I wanted to see if there were any correlations between gender and any of the four preference scales.
The phi coefficient was computed between all pairs (this coefficient is the most appropriate correlation to compute between two dichotomous variables). Here is the correlation matrix:
First, it’s important to note how things were coded.
Males = 1, Females = 0
Extraversion = 1, Introversion = 0
Sensing = 1, Intuiting = 0
Feeling = 1, Thinking = 0
Perceiving = 1, Judging = 0
So what does all this mean? Well, pretty much nothing, statistically-speaking. The only two significant correlations were between gender and Perceiving/Judging and Sensing/Intuiting and Perceiving/Judging. From the coding, the first significant correlation means that in the sample, there’s a tendency for males to score higher on Perceiving than Judging, and for females to score higher on Judging. The second significant correlation means that in the sample, there’s a tendency for those who score high on Feeling tend to score high on Perceiving, and a tendency for those to score high on Thinking to score high on Judging.
The rest of the correlations were non-significant, but they’re still interesting to look at. There’s a positive correlation between being female and scoring high on Extraversion, There’s a correlation between being male and scoring high on Feeling, and there’s a very, very weak correlation between Feeling/Thinking and Extraversion/Introversion.
Woo stats! Take the test, too, it’s pretty cool.
Today’s song: Beautiful Life by Ace of Base
Exactly how does one go about kissing the rain?
So as probably none of you know, I recently came into contact with an old friend from 7th grade. It was weird—early last week he randomly popped into my head and I thought, “I wonder what ever happened to Ross?” and then last Thursday I get a call from my dad—Ross had called him (sometimes it’s a good thing when at least someone in my family is able to keep the same phone number for more than 5 years) and dad gave me all his contact info.
Now we’re friends on Facebook and we talk on Messenger on occasion. He’s pretty much exactly the same, which is good to know ‘cause he was blind and insane and really awesome to hang with in 7th grade and is apparently still blind and insane and really awesome to IM.
Apparently I’m still the same, too. We spent the other night talking about the old Knowledge Bowl competitions we “participated in” solely because it meant we got to go to McDonald’s and then goof around on the bus to the competition/at the competition/on the bus back from the competition.
A few more memories passed between us, and then he said this: “You never seemed very happy, still breaks my heart to think about.”
This kinda surprised me. Really? I didn’t ever seem very happy? Back in junior high, when all I had to worry about was stalking Patrick learning how to type and not getting my thumb sawed off in shop class?
I find that very…disconcerting.
Do those of you who know me now find me unhappy? Am I like this harbinger of depression or something? ‘Cause I certainly don’t remember being Emo Central in 7th grade (that was 8th grade, but I was on meds and they killed my soul, among other things) but apparently that’s how I came across. There’s a lot more shit going down in my life right now than there was back then, but aside from the occasional “I HATE MY LIFE” blog—and let’s be honest, who doesn’t have those every once and awhile?—I don’t think I’m all that unhappy-sounding.
Meh. Probably overanalyzing it. I’M GONNA GO WRITE DEPRESSING POETRY NOW WOO!
Today’s song: Vancouver City (featuring Linda Ganzini) by Innerlife Project
This Week’s Science Blog: Microwaves – The Answer to Everything
I knew about the microwave background “noise,” but this video describes it and the paradox very well.
I also realize that I missed last week’s science blog, so here’s a link to the Internet Encyclopedia of Science. I know it’s no substitute for one of my obnoxious science reviews, but it’s certainly a lot more useful.
I LOVE YOU ALL!
RED BULL!!
Today’s song: Touch the Sky (Original) by Iambic
The Periodic Table of Academic Disciplines
Alternate title: Claudia’s Bored

(Click to enlarge!)
(Source for list of disciplines/categories)
YES I KNOW it’s not exactly like the actual periodic table groupings, but I gave it my best shot. I tried to keep the general “this is how things are organized” patterns, but some of the disciplines just didn’t fit in anywhere else (anatomy, I’m looking at you). I didn’t keep the P-, S-, D-, F-blocks ‘cause they didn’t work out in terms of layout, but I kind of made my own blocks instead (take THAT, Mendeleev!). The groups, however, are still sorta there.
Group 1: Formal sciences, applied
Group 2: Formal sciences, more theory-based
Group 3: Physics
Group 4: Physics-related stuff
Group 5: More specific physics
Group 6: Physics and space-related stuff
Group 7: Chemistry and biology
Group 8: More specific biology
Group 9: Specific types of organism-based biology
Group 10: Earth-related stuff
Group 11: Climate-related stuff
Group 12: More earth (ground)-related stuff
Group 13: Applied sciences that don’t fit anywhere else
Group 14: Arts (and marketing and accounting)
Group 15: More arts
Group 16: More arts
Group 17: Written arts
Group 18: Humanities
As for the colors, they’re more related to the blocks I guess. And the periods generally go from most fundamental/basic/theoretical (at the top) to the more applied (near the bottom of the columns).
Yeah.
I love Red Bull.
Today’s song: Alejandro by Lady Gaga (another “why the hell didn’t I have this song yet?” day)
The Four Corporations of the Apocalypse

I’m amazed by large corporations. I don’t know why. Maybe it’s because the larger companies get, the more fiercely they seem to push for their right to be the largest company, especially when they have to muscle out some other company for top spot. I find it amusing, interesting, and frightening all at the same time.
Due to the insane technological advances we’ve been making in the past decade, along with help from the glorious, glorious internet, we’ve been able to witness the birth of mega corporations that are able to grow to tremendous sizes and pretty much envelop everything they touch. And when they decide to merge, we’re all in trouble.
Oh come on, you know which ones I mean…
Microsoft (area of dominance: computing)
Not only does Microsoft (in my opinion, at least) pretty much own the computing sector with their PCs, they’ve also got quite a monopoly on software (Microsoft Office, anyone?), plus Internet Explorer, Zune, Windows Media, Windows Live (including Hotmail and Messenger), and the Xbox 360.
It’s probably the weakest of the four corporate giants as I see them, but it’s still got a pretty strong hold on things when you think about it. Hell, I typed this out in Microsoft Word and uploaded it in IE*. I guess the reason it seems weak is because it’s not expanding at the rate of the other corporations I’ve listed.
Speaking of expansion…
Apple (area of dominance: portable media)
Remember that time where Apple only made those dorky computers? Haha, yeah. Nostalgia. Now there are IPHONES EVERYWHERE. Perhaps you read my blog about my adventure in the Apple store. If not, go read it, slacker! here’s the summary: people are psycho for Apple products. The company is rapidly gaining ground in the portable media sector.
– Music.
– Phones.
– Phones that also play music.
– Wi-Fi access in small electronics.
– Wi-Fi access in small electronics that also act as phones and play music
– Whatever the hell the iPad is.
– But wait! A newer version of the Wi-Fi/phone/music thing!
You get the idea.
Apple has pretty much taken over the “check out this electronic doohickey I’m carrying!” area, and it doesn’t show any signs of slowing up. Pair this with the grip it’s got on the music sales industry via iTunes and you’re looking at one powerful company.
Facebook (area of dominance: personal information)
The king of personal information, Facebook as it stands right now is quite frightening. It’s not Big Brother we have to worry about, but each other, now that we’re able to pretty much list everything down to our genetic code on a social networking site. The worst part about it is how addicting it is. I’m not ashamed (though I probably should be) that I went through a freaky little withdrawal stage when I shut down my Facebook account for a few months back in May (April? Whenever), and was pretty much fully hooked on it again when I came back. Despite all the privacy issues Facebook’s having right now, I don’t think the number of people using the site will decrease by any significant amount anytime soon, thus leaving those of us in Facebook Land a good population in which to search and stalk.
Google (area of dominance: general information)
Last but certainly not least is Google. Google is terrifying.
Google will own the world in approximately seven more years.
In a decade, “Googling” will no longer just be a word for “searching via google.com” but will be a euphemism for all sorts of other things (possibly dirty things). In twenty years, we’ll have street views of Alpha Centauri.
Can you tell this company frightens me?
I guess if you name your company after something as big as a googol, you’re pretty much destined to be of the mindset to want to expand as much as possible. Their getting their hands on YouTube was the final “oh crap!” moment for me, now I’m just waiting for the blue, red, yellow and green takeover. Or should I say takeooooooooooooover.
Paranoia? Perhaps. But I’m waiting for the day Google decides to merge with Apple, they conquer Facebook, and Microsoft decides to join in just because. Then we’re screwed.
*Anyone who gives me browser choice crap is invited to come over and count the number of times Firefox crashes when I use it. That browser and I don’t get along, I like IE best, shut the hell up.
Today’s song: Protection (Sirius Mo Radio Edit) by Ben Mono
Blog #1,592: A Survey
Because sometimes you just have to do a survey.
Will you be dating in six months?
Doubtful.
Who is the last person to send you a message on text?
Translink (I’d texted to get the bus time)
Are you jealous of anyone right now?
Nah.
Hold hands with anyone last night?
Pfft.
What will you be doing tomorrow?
Going to campus, doing research all day, trying to keep my soul from self-destructing…you know, the usual.
Why did you last cry?
I pissed myself off.
When will your next kiss take place?
That’s an excellent question.
You have to get a facial piercing, what do you get?
Nose ring! I’ve wanted one for awhile.
When was the last time someone of the opposite sex gave you a hug?
Um…July?
Is anything wrong?
Yes.
Could you cry right now?
Nah. I’m watching Ocean’s 11 and Annabelle’s in my lap.
Are you someone who worries too often?
I think there was one day back in high school where I remember being worry-free.
Who was the last person you took a picture with?
July, with Matt, Maggie, and Rebeca. CAUSE WE’RE AWESOME!
Was yesterday better than today?
Yeah.
What’s the very first thing you do when you wake up?
Turn the alarm off.
Do you ever think about stuff and start crying?
Haha, do you know me?
Do you judge people you don’t know?
I certainly try not to.
Are you happy with the way things are going?
No.
Where did you go in a car last?
To Safeway in the Oakridge mall.
Are you open about your feelings, or closed off?
It depends on the day/feeling.
Who was the last person you were in a car with besides family?
Sean and Megan.
Have you ever felt replaced?
Yeah, a few times.
Should you be doing something else right now?
Nope. It’s scheduled down time now.
How were you feeling last night?
So-so.
Are you different from how you were a year ago?
Hahaha, oh yes.
If the person you fell hardest for died today, how would you feel?
That would be pretty horrible; I don’t really know how I’d handle that.
Do you like your life as of now?
Bah.
Is there a person of the opposite sex who means a lot to you?
Several.
Do you wish someone would call or text you right now?
It’s probably too late for anyone to do so, if they even wanted to.
Do you like surprises?
Sure.
When was the last time you laughed hard?
Yesterday. I stumbled upon Uncyclopedia’s Vancouver article. Read it, it’s GREAT.
Who was the last person you laid in a bed with?
Hahaha, that’s a private matter.
Who made you smile today?
The guy at the rec center mouthing along to Bad Romance.
What made you sad today?
Me.
What keeps you up at night?
See above.
What did you do yesterday?
Research, went to Multivariate Analysis, rode the bus, read, wrote, made an R tutorial video and subsequently lost my voice.
Who was the last person you had a deep conversation with?
Sean?
Have you heard a song that reminds you of someone today?
Not today.
Do you hate the last girl you had a conversation with?
My mom? Hahaha, no.
When is your birthday?
February 2nd.
Last person you held hands with, do they mean anything to you?
I haven’t held hands with someone in a long time.
Does it take a lot for you to cry?
Nope.
Name five things that are next to you.
Water bottle, Xbox controller, journal, pen, comb, headphones.
Are you shy?
Socially anxious is a better phrase for it.
Are you listening to music right now?
Nope, TV.
Do you have any weird inside jokes?
Hahaha, my life is a weird inside joke.
What is the last thing someone bought you?
Groceries.
Who was the last person you talked to last night in person before bed?
Uh…my office mate?
Have you made a mistake this past week?
Indeed.
What were you doing at 4am this morning?
Probably having weird dreams.
What’s something that can always make you feel better?
Sleepyhead.
Do you have both a loud side and a quiet side?
YES (that was the loud side).
Are you a stressed out person?
Hahahahahahaha.
I’ll bet you miss someone right now?
I miss a lot of someones right now.
Do you think someone is thinking about you right now?
Doubt it.
Who is the last person you high-fived?
Matt?
Do you remember your dreams?
Sometimes.
Is it okay if you kiss people when you are single?
Sure?
Would you date someone right now if they asked you?
Depends on who would ask me.
Are your parents over the age of 40?
Indeed.
Are you easily amused?
Oh god yes.
Does the last person who put their arm around you mean anything to you?
Yes.
Do you miss your past?
Parts of it.
Were you single on your last birthday?
Yep. Back to the old standard.
Do you think you’re wasting your time on the person you like?
Oh, that’s a complicated question.
Do you have a reason to smile right now?
Sorta.
How are you feeling?
Sick-ish.
Is there something you’re not looking forward to?
The next year.
What is your favorite thing to shop for?
Trinkets.
Do you usually listen to your parents?
Yup.
Ever dyed your hair?
Twice.
Would you rather own a snake or a rabbit?
Rabbit.
What is the last non-alcoholic beverage you had?
Water.
Do you tell your parents everything?
Nope.
How many hours did you sleep last night?
Five?
Today’s song: Breathless by The Corrs
I don’t care that I’m 22 years old, I think Adventure Time is genius.
Back when I was in Moscow over the summer I found this show while channel surfing through the 300+ channels my dad has. I kept watching because the art/humor style was similar to that ofTom Deslongchamp (who is super awesome), but I kept watching because it’s freaking amazing.
Why? Well for one, they did this:
(if you don’t get it, go here)
Here’s one of the episodes, ‘cause it’s just fantastic.
Yay. :D
Today’s song: Runaway by The Corrs
Every Time You Misinterpret a Confidence Interval, God Kills a Statistician
HI AGAIN!
So I’m liking WordPress. I’m liking it enough that I’ve started an entirely new blog dedicated to R. It’s right here. It, unlike this one, will actually kind of look professional and will not consist of my daily ramblings. Rather, it will contain info on using R as well as little tricks, ’cause we all know R can be a pain sometimes.
OH YEAH, I also found an old high school project of mine that revolved around quite a nice little dataset. My “statistics” back then were horrendous, so I’m thinking I might re-do the whole thing and post it up here. Plus my graphs were made in Excel and look terrible. That has to be remedied.
WOO SHORT BLOGS! School starts tomorrow, I’m stressed, deal with it.















