Nonparametric test for event study

[Sang-Bum Park]

Does Stata have some commands to test normality assumptions for using event study? If so, what commands are specific for the nonparametric test?

[Carlo Lazzaro]

Sang-Bum:
welcome to the list.
Could you please post further details about what you're after? Thanks.

[Marcos Almeida]

Carlo is absolutely right. It is difficult to get the gist of this query. Please follow Carlo's recommendations if you want to increase the odds of your query being clarified.

That said, and just hazarding a guess in a Sunday morning. I will assume Sang-Bum might be eventually talking about the semi-parametric Cox regression and its assumption of proportionality of hazards. This being the case, such assumption could be checked, among several options, by typing either - stphtest, detail - or - stphplot, by(group) - for a visual display.

With regards to the tests of equality of survivor functions (such as log-rank,Wilcoxon, Tarone-Ware, Peto-Peto-Prentice and Fleming-Harrignton), they are basically related to the use - or not - of weights so as to adjust for earlier failures, later failures or both, depending on the selected scheme. Here, the trigger to the "hypothesis" may well come from the presentation of Kaplan-Meier curves.

Hopefully that helps.

Best,

Marcos

[Carlo Lazzaro]

Sang_Bum:
as an aside to Marcos' excellent insight; I would recommend you to take a comprehensive tour through -help st- and related entries in Stata 13.1 .pdf manual, where you will find a thorough coverage of all the stuff mentioned by Marcos (and much more else).

[Sang-Bum Park]

I am very grateful for your (Carlo Lazzaro and Marcos Almeida) extremely kind answers. I am really sorry for my late response. As a novice, I have not visited the site regularly. Now I have seen your answers, which were all very helpful. In this question, I asked about event study, which is often used to analyze with stock information in the field of financial management. Although Marcos Almeida link my question to event history analysis, I agree with Carlo Lazzaro that his answers are insightful and very intriguing. Thank you all very much.

[Marcos Almeida]

Sang-Bum Park Thank you for your feedback.