My research productivity was way down since moving to Unknown University. Getting set up at a new school always takes more time than I expect, so the fall wasn't that productive. And the first part of the Spring semester was tied up with prepping for the CFA classes I've been teaching, which took far longer than I'd planned on.
But now that that's over, I'm back to research. So far this semester, I've submitted two papers to journals and gotten papers accepted at three conferences (granted, they're smaller regional conferences rather than the national ones, but it's a start). And this week, I should finish up a third (smaller) paper that's been in progress for FAR too long.
Now that I've cleared the deck of my older papers, I get to work on some newer ideas (caution: nerd alert ahead).
Without revealing too much information (after all, I am blogging pseudonymously), I've been working on a project lately that requires me to make pairwise comparisons between each and every security analyst who makes forecasts in a given two-year period. This results in over 25 million pairwise comparisons (and that's just for a two-year period). If you think this is a huge amount of data, you're right - it is.
But the truly amazing thing is how easily my computer can handle this. My setup is pretty basic - I'm running SAS on a 6 month old desktop machine with 2 gigs of ram. The program runs through one iteration of the analysis in less than two minutes. In the grad school days of yore, it would require a day or two on the mainframe, and I'd have to break the job into multiple parts to avoid crashing the system.
Life in the time since the meteor hit and the dinosaurs died is good.
No comments:
Post a Comment