Abstract
With an adaptive partition procedure, we can partition a “time course” into consecutive non-overlapped intervals such that the population means/proportions of the observations in two adjacent intervals are significantly different at a given level . However, the widely used recursive combination or partition procedures do not guarantee a global optimization. We propose a modified dynamic programming algorithm to achieve a global optimization. Our method can provide consistent estimation results. In a comprehensive simulation study, our method shows an improved performance when it is compared to the recursive combination/partition procedures. In practice, can be determined based on a cross-validation procedure. As an application, we consider the well-known Pima Indian Diabetes data. We explore the relationship among the diabetes risk and several important variables including the plasma glucose concentration, body mass index and age.
Highlights
Time-course data analysis can be common in biomedical studies
We explore the relationship among the diabetes risk and several important variables including the plasma glucose concentration, body mass index (BMI) and age
Since we know the true parameters for simulations, we choose the aC value that minimizes the overall mean squared error (MSE)
Summary
Time-course data analysis can be common in biomedical studies. A ‘‘time course’’ is not necessarily only a certain period of time in the study. It can be patients’ age records or biomarkers’ chromosomal locations. A general ‘‘time’’ variable can be a predictor with continuous or ordinal values. When we analyze a response variable (binary, continuous, etc.), it is usually necessary to incorporate the information from this predictor. We focus on a nonparametric type of analysis of time-course data: the whole time course is partitioned into consecutive non-overlapped intervals such that the response observations are similar in the same block but different in adjacent blocks. The partition of time course is the detection of change-points. For the detection of multiple change-points, since there are many unknown parameters like the number of change-points, the locations of change-points and the population means/ proportions in each block, it still remains a difficult problem [2]
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have