Maximum Likelihood Estimation

[Rasmus Thomsen]

Hey

I am supposed to estimate the three parameters (i.e. lambda0, lambda1 and delta) in the following equation by maximum likelihood:

I am not so familiar with maximum likelihood in Stata but as far as i understand i should be able to estimate the parameters by using the mlexp-command. However, I keep having the error that stata could not find feasible values r(491).
Is it not possible to use the mlexp-command when i want to estimate this equation? Do i have to use the more complicated ml-command?
I am quite sure that i have typed the equation in the right way, so I am suspecting that the mlexp-command is not sufficient to estimate the parameters from equation 3.16.

Below is my code.
Notice that h and r is just the maximum and minimum value of the wage variable.

global brok "((756.66 - wage + {delta}^2 * ((wage - 15.42)/({delta} + {lambda})^2)^0.5/(756.66-15.42)^0.5)"

mlexp   (log({gamma}) - {gamma} * duration + log({lambda} + 2 * {delta}) - log ({delta} + {lambda}) ///
             - 0.5 * log(756.66 - wage + {delta}^2 * ((wage - 15.42)/({delta} + {lambda})^2)) - 0.5 * log(756.66 - 15.42) ///
- a * log((1-c) * {delta} ///
+ c * (({delta} + {lambda}) * ( $brok ) * {delta})) ///
- ({delta} + {lambda}) * ($brok) * job), variables(job c a wage duration)

[Phil Bromiley]

You didn't get a quick answer. You'll increase your chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output and sample data using dataex. When a problem reflects the interaction of a model and data, it is hard to diagnose without data.

In general, when maximum likelihood doesn't run, it is often good to start with a simplified model and then work up. If this is a course assignment, remember we do not help with homework.