Figure 1
From: Longitudinal data analysis for rare variants detection with penalized quadratic inference function
Performance of the pQIF for different sample sizes and different dimensions. (a) p = 20, (b) p = 40. The horizontal axis represents the variables, where 1 represents covariate age, 2–4 represent the three gene variables (MAP4, TNN, and NRF1) when p = 20, and 2–6 represent the five gene variables (MAP4, TNN, LEPR, FLT3, and NRF1) when p = 40, others represent the noise variables. The true and working correlation structures were set as AR(1). The title of each subfigure (e.g., “n = 142” in the top left panel) refers to the sample size. Since the pQIF did not converge well for n = 142, p = 40 in some simulations runs, the estimation results were not listed in the figure.
