After a long bout of a lingering cold/cough/congestion which continues to haunt me, I decided that enough was enough and that I needed to get back on track with my P90X plan.  When my infection initially hit, I was left breathless, fatigued and hacking.  For those of us doing P90X, that’s the norm, right?    Only this time it was not the worked out feeling I was experiencing.  My body simply could not perform.  So, for many frustrated days I was unable to stay on track with a true 90 day plan.  I think I’m probably about a week off, so I could label my trek as P97X.  Regardless, I feel like my body is at the point where symptoms are starting to subside to the point that I can perform and hopefully push this junk completely out of my system.
Last night was my last time at the Chest, Shoulders and Triceps workout.  You start this program in the 2nd phase during Week 5.  I’ll miss this workout because it’s really a fun one that blasts the upper body and really challenges you. 

The first graph you see above I listed all the exercises that you are to do using your bodyweight only.  So you’ll see many push ups, etc.  Upon review of the data, I saw great progress in most areas.  MOST…    I’ve done more tri-rises in a different workout, but the time you are allotted kept my number from going as high as I’ve actually done (which is 23).  You’ll notice the big Goose Eggs on One Arm Pushups.  More power to Jack Palance for doing these, but by the time you get to these in the workout plan, my triceps are toasted and the body is simply not willing to put out any effort for these (even on my knees).  Overall, including all reps, I improved 63% from start to finish. 

On the second graph you’ll see I now list all exercises using weights.  To attempt to figure improvement, I introduced a concept I’ve learned called “Tonnage” where you multiply your weights by your reps to get a figure that you can compare to the back end.  This concept worked to show improvement in all but a few areas.  For Pour Flys, Front to Back Tricep Extensions and DB Cross-Body Blows, there was no improvement shown due to the tonnage model because at Week 5 I was doing high reps with low weights.  By Week 12 I was doing higher weights with less reps.  So, while my strength was up, this is where the tonnage model needs an asterisk by it and shows once again the difficulty comparing actual effort in these cases.  Regardless, I showed a total strength increase of 42%.
Whether flawed or not, taking 63% and 42% and averaging them out, I’m a little over 50% stronger than I was a few weeks ago.  I don’t know about you, but that does bring a smile to my face as it shows validity once again to the outcomes P90X can produce.  The numbers don’t lie.
Ready to allow P90X to change you?  If so, I would enjoy being your Coach (free of charge) to motivate you and guide you and advise you on your way. 

Click on this LINK to get started or to learn more!

Yikes.  About a third of Americans are already obese, and the rate of increase is holding steady.  Whether you’re more worried about the cost in human lives or to our economy, it’s clearly time to take this problem seriously.  Next time: more on ways to tackle it.

Just wanted to put a couple of charts up that give a pretty decent view of Albert’s offense through the years.  First, his wOBA by components graph (note these were computed with each year’s coefficients)

That’s pretty consistent, with only one year that tickles .400 or below.  And for another look, the same chart from a percentage view

This gives a better view into the relative way he’s amassed his offensive production.  Wow he got a lot of his value from the HR in 2006…  anyway, just thought these graphs were interesting given the occasion of his 3rd MVP.

My plan is to get my simulation back out now that I have my defensive projections to input in there, so look for some comparisons of different potential roster constructions.  If you have any you’d specifically like to see drop them in the comments and I’ll see what I can do.

It was late and rushed tonight, but I did manage to do some math with the kids. Actually, A was throwing fits, so I put her to bed. I had B working on the fact that if you have a connected graph on 5 vertices then there is a connected subgraph on 4. He worked out this conjecture himself, but doesn’t have any justification for it. I’m thinking he should develop the idea of a spanning tree, but I’m not sure how to nudge him in that direction. I guess I’ll have to think up an intermediate problem.

People’s attitudes to and belief in climate change. (Source: TimesOnline street survey)

People’s attitudes to and belief in who was responsible for 9/11. (Source: WorldPublicOpinion.org)

People’s attitudes to and belief in alien visitation. (Source: LIFE magazine poll, 2000)

| Gabriel |

I’ve tried to learn R a few times but the syntax has always been opaque to my Stata-centric mind. I think the trick is to realize that two of the key differences are that:

• whereas handles only come up in a few Stata commands (e.g., file, postfile, log), they are very important in R, what with all the “<-” statements
• in R there’s a much fuzzier line between commands and functions than in Stata. What I mean by this is both the superficial thing of all the parentheses and also the more substantive issue that often you don’t put them one to a line and they just do something (like Stata commands) but you usually put them many to a line and feed them into something else (like Stata functions). Related to this is that the typical Stata line has the syntax “verb object, adverb” whereas the typical R line has the syntax “object <- verb(object2, adverb)”

The two combine in an obvious way with something as simple as opening a dataset, which is just use file in Stata but is filehandle <- read.table(“file”) in R, that is, there’s not a read.table() command but a read.table() function and you feed this function to a handle. (And people say R isn’t intuitive!)

At least I think that’s a good way to think about the basic syntax — I suck at R and I really could be totally wrong about this. (Pierre or Kieran please correct me).

Anyway, I wrote my first useful R file the other day. It reads my Pajek formatted network data on top 40 radio stations and does a graph.

# File-Name: testgraph.R
# Date: 2009-11-20
# Author: Gabriel Rossman
# Purpose: graph CHR station network
# Data Used: ties_bounded.net
# Packages Used: igraph
library(igraph)
setwd("~/Documents/Sjt/radio/survey")
chrnet <- read.graph("ties.net", c("pajek"))
pdf("~/Documents/book/images/chrnetworkbounded.pdf")
plot.igraph(chrnet, layout=layout.fruchterman.reingold, vertex.size=2, vertex.label=NA, vertex.color="red", edge.color="gray20", edge.arrow.size=0.3, margin=0)
dev.off()

The weird thing is that it works fine in R.app but breaks when I try to R run from the Terminal, regardless of whether I try to do it all in one line or first invoke R and then feed it the script. [Update: the issue is a 32 bit library and 64 bit R, the simple solution is to invoke "R32" rather than just plain "R". see the comments for details]. Here’s a session with both problems:

gabriel-rossmans-macbook-2:~ rossman$ Rscript ~/Documents/book/stata/testgraph.R
Error in dyn.load(file, DLLpath = DLLpath, ...) :
unable to load shared library '/Library/Frameworks/R.framework/Resources/library/igraph/libs/x86_64/igraph.so':
dlopen(/Library/Frameworks/R.framework/Resources/library/igraph/libs/x86_64/igraph.so, 10): Symbol not found: ___gmpz_clear
Referenced from: /Library/Frameworks/R.framework/Resources/library/igraph/libs/x86_64/igraph.so
Expected in: dynamic lookup

Error : .onLoad failed in 'loadNamespace' for 'igraph'
Error: package/namespace load failed for 'igraph'
Execution halted
gabriel-rossmans-macbook-2:~ rossman$ R 'source("~/Documents/book/stata/testgraph.R")'
ARGUMENT 'source("~/Documents/book/stata/testgraph.R")' __ignored__

R version 2.10.0 (2009-10-26)
Copyright (C) 2009 The R Foundation for Statistical Computing
ISBN 3-900051-07-0

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.

Natural language support but running in an English locale

R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.

Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.

> source("~/Documents/book/stata/testgraph.R")
Error in dyn.load(file, DLLpath = DLLpath, ...) :
unable to load shared library '/Library/Frameworks/R.framework/Resources/library/igraph/libs/x86_64/igraph.so':
dlopen(/Library/Frameworks/R.framework/Resources/library/igraph/libs/x86_64/igraph.so, 10): Symbol not found: ___gmpz_clear
Referenced from: /Library/Frameworks/R.framework/Resources/library/igraph/libs/x86_64/igraph.so
Expected in: dynamic lookup

Error : .onLoad failed in 'loadNamespace' for 'igraph'
Error: package/namespace load failed for 'igraph'
>

The problem seems to be that R (terminal) can’t find the igraph library. This is weird because R.app has no trouble finding it. Furthermore, I get the same error even if I make sure igraph is installed directly from R (Terminal) in the same R session:

chooseCRANmirror(graphics = FALSE)
install.packages("igraph")
source("/Users/rossman/Documents/book/stata/testgraph.R")

I guess that’s another difference with Stata, StataConsole knows where the ado library is. I’d like to be able to use the Terminal mode for R as this would let me to reach my Nirvana-like goal of having a single script that does everything without any proximate human intervention. So I’ll just ask? How do I get R (Terminal) to run as reliably as R.app? Is this a naive question?

Or would it be better to try to feed R.app a “source” script from the command line? Much how like I can do this for Stata to launch a do-file into the Stata GUI
exec /Applications/Stata/StataMP.app/Contents/MacOS/stataMP ~/Documents/book/stata/import.do

Today I was working with a teacher using Photo Filtre to enhance images. We used inspiration from the artist Andy Warhol’s art. Here is an example showing one of the neat effects you can use! Photo Filtre is freeware – click here if you want to download it and try some of the filtres and colours yourself.

The University of Waterloo’s CEMC holds Math Circles in which they invite local grade 6-12 kids to come to the university one evening a week for math enrichment activities. In the past years I have participated for 2-3 weeks per year, running sessions on the topics of Game Theory and Conics (ellipses, parabolas, etc). This year my topic is Graphs! Here are six small graphs (you can ignore the numbers on the edges):

The first session, yesterday, was very enjoyable (at least to me, but it also seemed to get positive feedback from the other participants as well).

In my experience, it can be tricky to guess the appropriate level of discourse when preparing the lessons, mostly since I have no prior experience with most of the kids in the session. This caused me to “lose” about a 1/3 of my first session to covering background material: proof by induction. (It’s not really a “loss” because I would say that the time spent was still fun for me and beneficial for the students.)

In the course of developing the online notes to accompany the presentation, I wanted to see if there was a good resource online with an appropriate explanation of “proof by induction.” There are certainly hundreds of books including the subject, and given the size of the internet, one might conclude that there are thousands of articles online about induction? Well, it doesn’t quite seem to be that many. I found that Wikipedia had the best all-purpose description (with an example, and without getting too abstract) and there is another online at Carnegie Mellon University. I was hoping to find something good at the Art of Problem Solving website but their description was a little too terse. It looks like there is still some room on the internet for well-written mathematical enrichment material, I am glad the Math Circles content is available to help fill that hole.

An interesting graph from David Rosenberg:

The radius of a graph is the minimum eccentricity of all vertices [1].

What is “eccentricity”? The eccentricity of a vertex v is found by (i) obtaining the lengths of the shortest paths from vertex v to all other vertices and then (ii) finding the maximum of those shortest paths.  Thus the eccentricity of a vertex is the greatest distance you’d ever have to travel if you were at vertex v and wanted to go to another vertex [2].

The radius is found by (i) obtaining the eccentricities of all the vertices in a graph and then (ii) finding the smallest eccentricity.

The following (iGraph-based) R code can be used to find eccentricities, the radius, and the diameter (which happens to be defined as the maximum eccentricity of the graph, or the longest shortest path [3]):

graph.eccentricities <- function(graph, ...) {
  apply(shortest.paths(graph, ...), 1, max)
}

graph.radius <- function(graph, ...) {
  min(graph.eccentricities(graph, ...))
}
graph.diameter <- function(graph, ...) {
  max(graph.eccentricities(graph, ...))
}

[1] Weisstein, Eric W. “Graph Radius.” From MathWorld–A Wolfram Web Resource. http://mathworld.wolfram.com/GraphRadius.html

[2] Weisstein, Eric W. “Graph Eccentricity.” From MathWorld–A Wolfram Web Resource. http://mathworld.wolfram.com/GraphEccentricity.html

[3] Weisstein, Eric W. “Graph Diameter.” From MathWorld–A Wolfram Web Resource. http://mathworld.wolfram.com/GraphDiameter.html

Also called “average path length”, characteristic path length is the average length of the shortest paths between all nodes in a graph [1].  Using iGraph under R we can calculate this measure using the following code:

characteristic.path.length <- function(graph, ...) {
  n <- vcount(graph)
  sp <- (shortest.paths(graph, ...))
  sp <- sp[is.finite(sp)]
  if (length(sp)!=n*n) warning(paste("This graph is not connected;
  calculating average length of shortest paths between connected vertices.
  There were",n*n-length(sp),"unreachable pairs."))
  sum(sp)/(length(sp)-n)
}

I have tested this and it produces the same result as Pajek.  Note that unlike the definition in Lovejoy and Loch [1], for disconnected networks Pajek averages only the distances between connected pairs.

[1] William S. Lovejoy and Christoph H. Loch.  2003.  “Minimal and Maximal Characteristic Path Lengths in Connected Sociomatrices” in Social Networks 25.  Pages 333–347.

Yay! Some pictures of my actual work!! Docs can be cool too! Well, if you think graphs are cool… Which for the record, I do.

Even though I am still having another issue, I seem to have Dehydra working well enough to generate some docs. Basically ignoring the previous issue, I came across:

set -e;  \
for class in `cat graphtypes.list`; do \
dot -Tpng -o${class}-graph.png -Tcmapx -o${class}-graph.map ${class}-graph.gv; \
done
/bin/sh: line 2: dot: command not found
make: *** [classapi] Error 127

This happens to be a nice, soothing error at this point. I found the dot command and tried yum search "directed graphs" which pointed me to Graphviz.

A quick yum install graphviz later and no problem! Up above are a couple of the graphs generated for what I will be making for the MDC.

Following on from centrality normalisation are graph-wide centralisation measures.  These too were defined by Freeman.  The following are implementations in R.  They have been checked against test networks and match the values produced by Pajek.

degree.centralisation <- function(graph, ...) {
  deg <- degree(g, ...)
  n <- vcount(graph)
  (max(deg)*n - sum(deg)) / (n*n - 3*n +2)
}

betweenness.centralisation <- function(graph, ...) {
  bet <- betweenness(graph, ...)
  n <- vcount(graph)
  2 * (max(bet)*n - sum(bet)) / (n*n*n - 4*n*n + 5*n - 2)
}

closeness.centralisation <- function(graph, ...) {
  if (!is.connected(graph)) warning("This graph is disconnected; calculation\
  of closeness centrality on a disconnected graph is only possible because\
  igraph assumes a non-infinite distance (that is equal to the number of\
  vertices in the network) between two disconnected vertices.")
  clo <- closeness(g, ...)
  n <- vcount(graph)
  (max(clo)*n - sum(clo))*(2*n - 3) / (n*n - 3*n + 2)
}

There are a few graph centralisation measures that occur in iGraph in their raw form, but sometimes it would be nice if they were normalised.  The implementation of closeness is normalised, but degree- and betweenness-centrality are not.  Freeman’s paper [1] discusses how they should be normalised and the following functions are implementations in R.

[1] Linton C. Freeman.  1979.  “Centrality in Social Networks Conceptual Clarification” in Social Networks, 1 (1978/79).  Pages 215–239.

degree.norm <- function(graph, ...) {
  deg <- degree(g, ...)
  n <- vcount(graph)
  deg/(n-1)
}
betweenness.norm <- function(graph, ...) {
  bet <- betweenness(graph, ...)
  n <- vcount(graph)
  (2 * bet) / (n*n - 3*n + 2)
}

You have surely had the days when you meet that perfect person for the ‘dumb as a dodo’ badge. By ‘dumb’, i mean, THIS dumb.This is probably what ran through during that conversation (not necessarily in the same order).

1. “Uh.. what?”
2. “Wow.. THis is amusing.. is she/he really that dumb?”
3. “okay. enough.. stop..”
4. “stop stop … .STOPPPPPP”
5. *Frustration.. usually characterized by grinding of the teeth, shuffling of the feet, checking the time, and grinning non-commitally.
6. the next ten min are characterised by, ‘Attempts to strangle self with computer wire, puncture heart with safety pin..’

And then finally, you wrench yourself out of there. After the ‘phews’ and the ‘happada-s’, you (sometimes) sit back and wonder whom to blame. Is it god? or should we simply blame Darwin.You see, Darwin (some 300 years ago) proposed the theory of natural selection, to the layman – ’survival of the fittest’. The theory states exactly what it spells. A species survives if it possesses ’superior’ traits in comparison to another species. Now that can apply to humans as well and especially to dumb people. Easy solution? Do what Darwin would? Pray that the dumb people cease to exist around your circle of influence and get on with your life. NO.

I say (and i can prove it), that the theory of natural selection breaks down here. IT WILL NOT WORK. YOU WILL NOT BE BETTER OFF IF YOU GET RID OF THE DUMB PEOPLE.Heres why,

Proof: Now I shall turn to the old trusty gospels of statistics.. The law of large numbers and the central limit theorem. The central limit theorem says that any set of random variables (peoples IQ in my case) converge to a normal distribution. What does that mean? Not much until you factor in that, the dumb people take up the left skew of the normal curve (refer fig).

YOu dont need to be a statistician to do the rest. The dumb folks are the ones who are pulling the mean leftward. Now why is that good for you? Thats because you are at the right end of the curve. The dumber the people you meet are, the smarter you will become in society. Having not so smart people around is good for you, my friends. it is.. So, preserve them… for they will highlight your importance to the world. I leave you with this video (and there are many more out there). Amen.

I’ve played quite well even though my results don’t show it. I’ve been ill and have played about half of my usual volume. On the whole i’m running fine at NL100 but not at NL200. That being said I should just give up on playing any stakes of Heads-Up game as it’s really not something I can claim consistency in. The variance is higher than 6max or full ring games anyhow so the sample size would have to be pretty large to even out. Not something I have time for right now.

Here are a few of the things that have got my attention recently whilst playing…

• Learning to be invulnerable. Plugging those leaks in my game before exploring newer, marginal areas of gameplay. Something that is a fundamental in ‘The Art of War’.
• Being calm and focused enough to ALWAYS remember poker is not a way to get rich quickly. A steady mind means less mistakes. Patience has been a big theme for me this month in conjunction with…
• Staying in the present to get into the zone. Too often I have drifted ‘off topic’ when the real enjoyment I get from playing is being perfectly right, in the moment. I have been taking the advice of a psychological coach to focus on relevant cues and nothing else. I can already see the improvement in my hand reading ability.

Being more patient has seen me make fewer mistakes at less expense. I’m much more willing now to play a whole session and be content with not getting the information I need to make a play on someone if the cost of getting that info in the short-term is too high. I’ve found it really is important to never pay too much for that information.

I heard a useful soundbite in a TV show recently. Something like “You don’t remember the ones you save; you only remember the ones you fail.” There’s a myriad of ways to interpret this into something useful in my life. Poker not the least. The people in my life especially. I don’t intend to fail any of these aspects and this just motivated me further to make sure I do my best and not settle for a weak effort.

On motivation, this game can be a trial on the mind to overcome the strains but, ethically or morally frowned on as I realise some motivations may be, each person has their own particular way of finding purpose. As much as I want to be the greatest at this…and I do, i’m feeling more motivated if i’m doing my best to win the most money as I can, not necessarily be the best I can. I’m certain they go together, but the structure of my thoughts in this way is proving more beneficial to me than searching for that champion spirit. It’s almost a release of external pressure as i’m no longer judging myself by an intangible but something I can get immediate feedback from.

An interesting snippet from this month was that I played a 19 hour session with a lunatic who had a VPIP of 98%, and another of 68%. Both at the same table, they blew about $2000+ in about 3 hours between them. My run of luck was so bad at first, and then I won it all back plus a little for my trouble. I captured it on Video and may post it one day. Honestly one of the craziest, highest intensity, pot swelling sessions i’ve EVER had. Funny stuff.

Anyway, here are the mid month graphs. I’ve added a Player summary to add a little understanding to the choppy graphs.

I got a haircut last night and this graph sums up how I feel about its look:

In its latest GBD report, Marc Faber used Gold as deflator of the nominal US GDP, effectively creating an index tracking the US GDP in gold term.

As we can see below, the results are pretty interesting:

The GDP / gold ratio is moving in long term waves or trends, which coincide pretty well with real economic growth (uptrends) and nominal growth (downtrends).

As an aside I also looking at the GDP/gold ratio versus the US unemployment rate:

It is not perfect, but high GDP / gold ratio tends to coincide with low unemployment rate.  It seems to make sense as high unemployment also means low GDP growth, high federal deficit and low interest rate… which can be seens as a good combination for gold.

Every volcano disaster movie from Volcano (1997) with Tommy Lee Jones to Dante’s Peak (1997) with Pierce Brosnan someone somewhere tries to out run a lava flow. Is this possible?

The answer is maybe. You will find everything you want to know about lava flows @ http://volcano.oregonstate.edu/

 On January 10,1977, at Nyiragongo lava sprang from the sides of the volcano moving at speeds up to 40 miles per hour (60 km/hr). About 70 people were killed.

 Measuring the temperature of lava. Photograph by R.L. Christiansen, U.S. Geological Survey, January 9, 1973.
The fastest Lava flows recorded were in Hawaiiin 1950 when Mauna Loa erupted. The lava traveled at 6 miles (10 kilometers) per hour through thick forest. But once the lava flows became established and good channels developed, the lava in the channels was flowing at up to 60 km/hour (97 kph).

Speed of average sprinter = 19 – 24 kph

Can you out run lava?

Work it out mathspigs. You are 300m ahead of the lava flow and you need to get up a hill to escape, which is 1000m away. The lava is flowing at 30 kph and you are running at 18 kph.

Will you make it??????

Look, a nice site, because it’s combine art with science.
Visualizations provided on Many Eyes range from the ordinary to the experimental. They are deliberately providing a wide array of possibilities. Click the picture for interactive visualizations and start exploring data using your own eyes!

As percentages, BNP vote share versus turnout in the Yorkshire and the Humber 2009 elections to the European Parliament.

Since the beginning of this rally, economic surprises and asset prices were moving in tandem, meaning that good “surprises” were lifting prices. Since late October, economic surprises have been mostly negative, but most asset prices have continued to rally.

As an example, here is the Citigroup economic surprise index versus S&P 500:

and versus the CRB index:

It is interesting to see that the nature of this rally is changing, now totally disregarding bad news. Could it be a sign that we are entering into the last leg of this rally, ie. the melt-up phase?