Ninety seven consecutive seconds of dendroclimatology consensus

Consensus cartoon

Advertisements

15 thoughts on “Ninety seven consecutive seconds of dendroclimatology consensus

  1. He he! Which reminds me, I always wanted a long white lab coat, because clearly it makes one smarter.

    P.S. I didn’t realize Homer had an invisibility shirt. That would be even cooler than a white lab coat.

    • Homer had all kinds of incredible tricks up his sleeves.

      The white lab coat is remarkably like the letters “PhD” after one’s name. Confers extraordinary scientific ability, if not the urge to grab and mix random chemicals that shouldn’t be grabbed and mixed.

  2. Still have the ol white lab coat… though not sure it would fit any longer (damn gravity!!) Also have the PhD after my name. So this should double up feelings of grandeur??

    Disabused myself of the notion long ago after observing so many other PhDs with questionable levels of ability [and quite honestly doubting my own attainment of any measure of supremacy]. I’ve taken to considering the degree more a measure of persistence than actual intelligence.

    If you can’t explain your research to a stranger outside your field, you still have work to do.

    Oh – IMHO: Homer’s greatest trick was wooing Marge. She’s the greatest.

    • Yep, you should be twice as smart Clem. Of course, twice as much brilliance will now be expected of you. You can start by telling me if the Tigers are going to win the Central. 🙂

  3. Here’s a thought… why don’t we wait until the middle of next week and then I’ll prognosticate on the Tiger’s chances. 🙂 Have to confess I’d not been paying attention to the AL Central – more of an NL Central guy. The Cardinals got hot last week while the Brewers were ice cold – so the Cards moved into the front spot. But the Cards just got swept by the Reds… so I’m not sure I want to place a crown in that division either.

    San Francisco could be a player in the NL, and the Tigers seem poised to be contenders on the other side of the fence. How’s that for not committing? There’s no blood on my lab coat.

    • Outstanding vacillation Clem. I’m expecting big things from you and your Ouija board next week, which I would imagine is a better bet than the Tiger bullpen.

    • So as the midweek baseball results are nearly upon us the time has come for me to reveal the predictions we are sure the whole of the baseball world awaits.

      For the many statistically inclined among us I thought I might offer a peak into the bowels of the complicated models developed to produce these insightful predictions. Using B code (an obvious subset of R used for Baseball model development) I’ve coded a Markov substring to populate a matrix of Bayesian priors that take into account win percentages amongst teams situated within divisions. Said matrix was subsequently parsed to eliminate non-normal eigenvectors of a second order so that comparison to an inverted matrix populated by Bayesian priors of win percentages from games played by teams from different divisions would not bias the node forward least squares estimates obtained in the first iteration of the model.

      Of course we still have more than a week of regular season baseball left to play, but as we stand at roughly 150 games in the book the confidence intervals associated with these models can be trusted due to orthogonal pre-test delineations of head-to-head matchups. [I can see I’ll need to work on the phrasing on this latter aspect of the methods discussion, but perhaps an example output from the model will help our audience see the logic employed].

      So – without further ado, our model has discretely predicted that in the American League, the teams from Boston, Minnesota, and Texas will NOT factor in the 2014 post season. In the National League the confidence intervals are not as small around our estimates, but it does seem fairly clear that the teams from Philadelphia, Chicago, and Colorado will also go home at the end of the regular season.

      Due to the lousy level of competitiveness in the American League East Baltimore has already clinched. I will take no credit for the model’s corresponding prediction of such… except perhaps to notice that the model does seem to have some level of value. Similarly the National League East has been dismal in terms of competition and Washington has already clinched there. Again the model correctly predicted this outcome, but worth note for our studies – the CI on the NL East prediction was larger than the respective interval estimate for the AL East prediction.

      Moving to predictions of some value (ie, facts not yet established) the model predicts with very high levels of confidence that wild card teams will NOT emerge from either the AL East or the NL East. And now moving on to@#)!@&%#_

      Ooops, I seem to have to have a glitch in the software. Will have to debug and get back to you on other predictions. But I can offer that it looks like Detroit can still take its destiny into its own hands. 🙂 🙂

    • This is outstanding work Clem and you do not appear to be a poofdah. You are hereby admitted as the newest faculty member of the Statistics Department at the University of Hullaballoo.

  4. Jim:
    A good chuckle.

    Maybe you should do a “David Letterman” style “Top 10 issues” with dendrochronology…(actually, that might also be pretty informative).

    • Oh come on Jim… Letterman and crew seem to come up with 10 things even when their subject matter might logically suggest only one or two worthy thoughts to list. [indeed, sometimes the mediocrity of listed thoughts makes them more humorous]

      So start us off with your two most devastating issues, and we’ll build a list from there.

  5. It would be irresponsible and an obvious violation of the rules to attempt #6… so by all means, leave it out. Bruce would have it no other way.

Have at it

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s