So, I asked my wife to review my first two blog posts about Monte Carlo methods, and at first all I could get out of here was, "It's a little dry." A fair criticism, at least of the second article, which did not discuss the history of the American nuclear weapons program. (I briefly considered titling the first article "Here Comes the Boom.")
Math, unfortunately, is kind of boring when you're talking about simple examples. But this blog shouldn't always be like that. I was aiming for precise and understandable, as opposed to glib, but I forgot to bring the funny. I'll also try to use more pretty pictures and charts; probability distributions are easier to visualize than to describe.
Just to give you an idea of what's to come, I'm thinking I will cover the following progression of subjects:
- My background in the subject, and how I came to be interested enough in it to start a software project.
- Use cases: a series of articles on different areas these methods could be used which are personally relevant to me, such as personal financial planning, business planning and project management, insurance risk management and a few others.
- A summary of the existing Monte Carlo tools that are out there, and why I think they're inadequate. This article may or may not get into the design parameters and goals for my own alternative.
- Some choices I've made up front, such as my choice of programming language and my choice of a name for the project (I'm not telling you now, that would spoil the suspense!)
- Deep dives into the various design issues, probably starting with some discussion of terminology and levels of analysis so I can use some defined terms without having to spend a paragraph explaining them every time I use them.
Any feedback you might have on use cases that are relevant to you, or other things you think should be on that list, would be most welcome.