Trend Analysis
- Trend analysis is a statistical approach to evaluate the linear or nonlinear relationships between two variables. When we talk about the secular trend, one of those variables is time.
- Usually, I tabulate the estimates by grouped time periods at first
- Then, use a regression model [log(rate)] to evaluate trends with calendar time
- Or, Age-Period-Cohort model [log(rate)] to evaluate calendar time trends accounting for age or cohort effects
- We can also use ANOVA (ANalysis Of VAriance) with contrast coefficients (polynomial contrasts, orthogonal polynomial coefficient)
- NCHS (2018, No. 179). National Center for Health Statistics Guidelines for Analysis of Trends in "Series 2. Data Evaluation and Methods Research"
- Yin (2016). Statistical Methods and Software Package for Medical Trend Analysis in Health Rate Review Process
- HealthKnowledge: Time-trend analysis, time series designs
- NCI Software: Joinpoint is statistical software for the analysis of trends using joinpoint models.
- Kim (2000). Permutation tests for joinpoint regression with applications to cancer rates
- Chandler (2011). Statistical methods for Trend Detection and Analysis in the Environmental Sciences
No comments:
Post a Comment