Risk Estimation for Large Software Development Projects

I don't have anything substantive to say about this. I just wanted to drop a placeholder since this is an interesting topic that should lend itself to quantitative analysis, be meaningful and be interesting to me. But would this be too far into economics instead of CS?