i want to fit and draw a regression line and label the line, like this:
reg y x
matrix coefs=e(b)
loc beta1hat=coefs[1,1]
loc beta0hat=coefs[1,2]
lfit y x, title("fitted line: y = `beta0hat' + `beta1hat' * x")
when i do this, i get values for the estimated coefficients displayed
with far more precision than i want. so i want to display only, say, 2
digits after the decimal for each macro. i wasn't able to figure out a
way to apply a display format to a macro, so i tried adding this code
before the -lfit- command:
loc beta1hat = int(`beta1hat'*100+0.5)/100
loc beta0hat = int(`beta0hat'*100+0.5)/100
this works most of the time, but sometimes i get a graph that has a
title like this:
"y = 7.84000000001 + 2.31 * x"
or
"y = 7.83999999999 + 2.31 * x"
i assume that the problem is the rounding doesn't work in binary all the
time. can anyone suggest a way around this? (obviously if i were
making a single graph, i could put the coefficients in the title by
hand, but this code is part of a larger program, so i'd like it
automated.)
thanks,
sean.
_____________________________
please note new contact information:
_____________________________
sean f. reardon
associate professor of education and (by courtesy) sociology
school of education
485 lasuen mall, #315
stanford university
stanford, ca 94305-3096
650.736.8517 (office phone)
650.725.7412 (office fax)
[email protected]
_____________________________
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
