Explore chapters and articles related to this topic
Hazards
Published in Peter Cummings, Analysis of Incidence Rates, 2019
The Cox model allows the underlying rates to change over time by slicing time into discrete windows; comparisons are made at each failure time. In the jargon of survival analysis, the baseline hazard is allowed to vary over time. This variation is apparent in Figure 25.2, where the hazards (instantaneous rates) of death at each failure time have been combined into a continuous curve using a kernel smoother. The hazard rate of death fluctuates up and down around the mean (constant hazard) values from the parametric exponential survival model (Figure 25.2) The results from both the parametric and the Cox models both produced proportional hazard curves; the hazard ratio was 0.77 at each point in time. The smoothed hazard curves from the Cox model show the initial rise in rates as events start to occur. Males and older persons had a higher rate of death; after an initial die off in these groups, the proportion of the survivors who were male or old started to swing up and down. This produced the oscillation of the Cox-model hazard curves after about 3 years.
Caregiver priorities for endpoints to evaluate treatments for Prader-Willi syndrome: a best–worst scaling
Published in Journal of Medical Economics, 2018
Jui-Hua Tsai, Ann O. Scheimann, Shawn E. McCandless, Theresa V. Strong, John F. P. Bridges
Figure 3 shows the result of the best–worst score mapped against the age of the PWS family member using kernel smoother for estimation. This figure demonstrates the variation of caregiver priorities throughout the natural history of the disorder. “Improves hyperphagia symptoms” (Hyperphagia) is consistently considered most important by caregivers caring for all age groups. “Decreases problems with anxiety” (Anxiety) is closely tied with hyperphagia and ranked high on the caregivers’ list. “Decreases overweight problems” (Obesity) rises in importance as the individual ages. “Improves intellectual function” (Intellect) drops in significance between the 20–40 age group, but was considered very important after the age of 42. The priority of “Decreases temper outbursts” (Temper outburst) slowly begins to decline when the individual reaches the 30+ age group. “Decreases skin picking problems” (Skin picking) and “Decreases gastrointestinal problems” (Gastrointestinal) were considered by all age groups as the least important of the benefits listed. The changes in priorities as the PWS individual ages were consistent with the free text feedback from the caregivers, where one caregiver stated, “the importance of the different categories changes as the child ages”.
Independent and synergistic effects of self- and public stigmas on quality of life of HIV-infected persons
Published in AIDS Care, 2018
Linlin Lindayani, Kusman Ibrahim, Jung-Der Wang, Nai-Ying Ko
The mean proportion of presence of self-stigma and public-stigma and changes in HRQOL scores reflecting with and without self-stigma and public-stigma were estimated using a kernel-type smoother (Hwang, Tsauo, & Wang, 1996) given by:kth time; ith patient’s assessment time; 1996). The dynamic changes in these proportions were analyzed using the iSQoL software (http://www.stat.sinica.edu.tw/isqol/). Due to the fact that the skewed duration distribution to date may influence the results generalized by the kernel smoother, we conducted a test of normal duration distribution after diagnosis of HIV. The results or normality testing showed that duration after diagnosis with HIV was normally distributed based on the nonsignificant Kolmogorov–Smirnov test.
A new look at the inverse Gaussian distribution with applications to insurance and economic data
Published in Journal of Applied Statistics, 2019
Motivated by these considerations, in Section 3.2.1, rIG distributions are adopted to propose a well-defined kernel smoother – that is an estimator satisfying all the fundamental properties of a PDF – with support 3.2.2.