I'm putting together an OLS monte carlo experiment that looks at alternative methods of modelling a certain constructed dependent variable. Is it necessary for me to artificially generate an "error term" for my constructed dependent variable?
2006-10-30 02:08:55 · 2 answers · asked by Nic 1 in Science & Mathematics ➔ Mathematics
It depends on the details of what you want to model, but in general you want to add an realistic error term.
See here for examples:
http://shazam.econ.ubc.ca/intro/index.html
2006-10-30 02:21:03 · answer #1 · answered by cordefr 7 · 0⤊ 0⤋
No. The error will appear in the distribution of your other variables. The distribution functions for your inputs already include the effects of errors, so that the final MC distribution will reflect the effects of those errors cumulatively.
2006-10-30 10:15:50 · answer #2 · answered by mattmedfet 3 · 0⤊ 0⤋