Review Reports

Round 1

Reviewer 1 Report

Review comments on “Proportional Hazard model and proportional odds model under dependent truncated data”

 

Please find the comments below:

1)    Abstract should be rewritten- improve the quality of English and grammar throughout the journal

2)    In the abstract please mention/highlight what is the novelty

3)    Use proper format of abbreviations!

4)    What author means by survival studies-change the word or elaborate properly

5)    Introduction part need to rewritten before stating about truncated variable, explain about truncated variable and why you want to focus on that. Make flow and give some background and comparison with other studies

6)    Carefully check the references styles. Please be consistent

7)     Use proper formula format and number it (for example….(i), …… (ii) ). Otherwise very hard to follow (before section 2.2 itself)

8)     Author mentioned about the previous study which is about quasi-independent- please give some background about this with proper references and make a flow about why you suddenly want to compare with quasi-independent -what is the relation

9)    Author mentioned that they employed a method by Chaieb et al., please mention what is new in your method and what is the difference you are making with previous studies

10) Equation 2 is not clearly explained more about the significance of the formula

11) Merge figure 1 and 2

12) “We found that the hazard of death in males was higher than females”- can you give some biological explanation why is it so

13) Conclusion is well written; however, it will be nice if author can mention specifically the advantage of using this method in future with different applications strategies other than AIDS

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper proposed the estimation procedures of the regression parameters under the proportional hazard model and proportional odds model on dependent truncated data.

It is well written paper and can be applied to many clinical data. I have one minor suggestion to the authors.

In a simulation study, authors used the Clayton copula and Frank copula for the dependence between X and Y  to generate the sample size with n = 100 (or 200) for each group. I suggest the author add gumbel copula simulated data generation to make sure your proposed model is working for many different asymmetric types of data.

If the authors follow my suggestion, I will be happy to accept the revised paper.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Dear,

Here is some points should be take under consideration in final version:

1.I can not found estimate of probabitity of selection

2. where's the effect of waiting time un real application 

3. The real data section must be retyping with more examples.

4. Where's estimate of hazard rate of real data.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf