Here is the code for today's MLE case.
Please remember to read the paper before our next session on 10/12:
Measuring and Testing the Impact of News on Volatility, Robert F. Engle and Victor K. Ng, The Journal of Finance, Vol. 48, No. 5 (Dec., 1993) (pp. 1749-1778).
Below is more general Pareto characterization of the left tail of a random variable.
Two of the new parameters come from the affine transformation (shift and scale), while the theta represents the parameter of the Bernoulli mixing variable.
I posted my discussion notes from out case about conditional expectation, including the observation about the connection with OLS regression.
I will use this forum for annoncements, and you are welcome to use it for public feedback. I encourage you to check it often. You may want to add it to your RSS feed.