Was just wondering if Kevin Schmit or Same Lazy Element could opine on something. I am trying to develop my own factor model for something that I want to trade on. Specifically, the risk of companies disappointing on guidance. I was wondering if I should be using a Cox-Proportional Hazard model by defining the hazard as being Disappointing/Not Disappointing and building time dependent covariates as such: Here are some questions: 1- Do you think the censoring is informed and that I should be making adjustments for this? If so, and I adjust for this, does the model output still have value? 2- Do you think I should start with a clean data set and then add some data points manually (for example, there are lots of info about company profiles which are not contained in databases, such as number of revenue streams and their nature). 3- I am perplexed by the lack of survival analysis used in building trading models. Why do you think that is? Isn't Survival analysis a nice way of handicapping the probabilities vs. the market implied probability density function? (I am comparing it to Multi variable regression or PCA which has a less intuitive interpretation for me). Thanks and sorry if I don't know what I'm talking about. I'm a beginner...but I really like the idea of being able to assess the baseline estimated probability of a trading catalyst.