This study compared the performance of several vertical scaling methods under various conditions. All methods tended to recover parameters well when sample size is larger and test is longer.
1. In figure 1, why the bias of the Mean/Mean method dropped dramatically when test length was 20? And increased when test length is 30?
2. In figure 4, what can be inferred from the performance of the Mean/Mean method as sample size getting larger?
3. As mentioned in the literature review, previous research did not come up with consistent results. What is the possible reason for that? Although this study stated that the results were consistent with some of the previous research, it did not explain why the results are similar with only some of the previous studies, and different from others. Therefore, I still don’t understand what makes the true differences between various methods.