A Semiparametric Tilt Optimality Model
Round 1
Reviewer 1 Report
The paper is well-written, understandable and presents interesting results. I think it can be published in its current form. The only remark is that in Table 1 in all the columns headed "ANOVA" this head should be capitalized.
Author Response
- Comment: The paper is well-written, understandable and presents interesting results. I think it can be published in its current form. The only remark is that in Table 1 in all the columns headed "ANOVA" this head should be capitalized.
Authors’ Response: We thank the reviewer for noting this capitalized issue. We now changed the Table 1 column names as ANOVA.
Manuscript change: Table 1, page 15
Reviewer 2 Report
1. When g(x) is known, the exponential tilt model (1.4) is the optimum model in the sense that it is the closest to g(x) among all probability distributions in C from (2.2) in KL-distance. Discuss.
2. Theorem 3 needs more explanations.
3. In simulation Section, please list an algorism to generate data.
4. Can we use the proposed approach for modelling over- and under-dispersed data?. Discuss.
5. If the data have extreme or outliers' observations, how can we apply the proposed approach?. Please discuss.
6. For various sample sizes in simulation Section, please discuss the results when n grows to infinity as an example n = 100, 200, 500, ....
7. In application Section, non-parametric plots like strip, violin, normal-QQ, and kernel should be included to discuss the behavior of data before analyzing.
8. Empirical plots for application section should be also plotted to prove the theoretical results.
Author Response
See attached word file.
Author Response File: 
Author Response.pdf
