A note about my job.

It has been expressed by some of my readers that they would appreciate a note about how I occupy myself, as related to my job. Perhaps they are just envious of my trips to the opera and my jaunts to the zoo--nonetheless, their demands have been noted. Hopefully you will find it amusing as well as informational.

As a consultant for the WHRC, I am mostly responsible to one Frank Merry. Frank was a member of my committee for my master's research, and also works on various projects with Greg Amacher, my major professor from Virginia Tech. Because Frank works directly with Greg, this means that I am also, indirectly, still responsible to Greg. Just to give you a taste of the two fine personalities that I work for, when I mentioned to Frank that I was doing some literature review and reading about cattle markets, he replied by saying that we should get the timber supply model out of the way, and...

"One thing you could do with cattle is go to the super market and learn the cuts."

Right. My job is diverse.

With respect to the timber supply, my job is to do the dirty work in the stats program: we use a program called limdep. The code for a piece of an input file tends to look something like this:

regress; lhs = ltagval; rhs = one, lyrs, lfam, lcapit, credit, lcdis,
lttime, ladco, lhrdry, lhrwet, lhire; hetero; keep=fitlagvl $

reject; fitlagvl=-999 $
dstat; rhs = vmpla, yronl, famsz, valci, , ctydis,
adco, pastar, tvalag, ltp2, wealth, exogin, ASST, BLOT, SQLOT, SQASST, DEFTIT $


?CALCULATING SHADOW WAGE
create; fmvpw=(1.65112579*exp(fitlagvl)/hrdayw) $
create; mvpw=(1.65112579*tvalag/hrdayw) $
dstat; rhs = fmvpw, mvpw $
create; lfmvpw = log(fmvpw) $

create; if(sdefor>0)sdefor=1; (else)sdefor=0 $
create; if(sforst>0)sforst=1; (else)sforst=0 $
create; if(sdefor=1 & sforst=1)newwood=3
;if(sforst=1 & sdefor=0)newwood=2
;if(sdefor=1 & sforst=0)newwood=1
;if(swood=0)newwood=0 $

logit; lhs = newwood; rhs= one, lyrs, lfam, lcapit, lwatr, credit, lexogin,
lcdis, lfor, lpast, lfmvpw, ltagval, ltp2, deftit; hetero; marginal effects;keep=psoldnew $


Basically, I run a bunch of regression models based upon some guidelines from Greg and Frank, and then I send them to Greg for feedback. 'Feedback' takes the following forms:

Greg Amacher, 10-27
1. why are wet and dry labor variables in the ag production function...is wet season labor important to production of crops? If dry season L is negative (not good), then maybe combine all labor into one variable.

2. Value of ag production should not be an explanatory variable in the sold wood equation and is correlated anyway with the shadow wage, which should be present in the sold wood equation. The presence of value of ag production is probably reducing the significance of the shadow wage in the regressions due to the correlation. The same goes for other wood sales regressions.

Greg Amacher (2) 10-27, Maria receives an email entitled:

"P.S. on my previous PS....I like the multinomial model....wow, about to have a heart attack from excitement...more inside"


Yesterday, Greg and Frank and I had an exchange about a diagram of my study area for the paper which we are hoping to publish from my master's thesis.

Frank: "Maria: have him (Paul) take out the communities and include only the surveyed communities, using dots for the icons"
Greg: "Don't agitate the dots damnit! Are you agitating the dots?"
Frank: "it just seems a bit messy at the moment. Did you have caffeine this morning?"
Maria: "That message certainly did sound caffeinated....Hope you haven't crossed back over to using the tools the rest of us need to be productive"
Greg: "Sorry, I thought you both knew about the FEDEX T.V. commericial where the guys are at a computer watching and talking to the "dots" (i.e., FEDEX trucks). Hey, I found it on the web, it's totally funny now given the Tapajos map thing. Here it is:
http://www.youtube.com/watch?v=19rVKy_pfFU
So, no caffeine but just a keen sense of humor on my end!"
Greg: "P.S. oh, it is a nextel commerical...."