From an application point of view this technique of modelling the trend and cyclical components can be perceived as a regression of the data Yt on periodic covariates, which are functions of time, while the stochastic dependence of the data, on its past including seasonality, is modelled by fitting SARIMA models to the residuals of this regression.
Using logistic regression, perceived exertion ≥4 (high exertion), referencing <4 (low-to-moderate exertion), was related to high activity of the trapezius muscle [OR 18 (95 % CI 2 143)], i.e., the odds for experiencing high exertion during work increased 18-fold for each percentage increase in time above 60%% MVC.
In addition, to measure the random variation of the risk perceptions in both groups we chose the SD of the residuals of the regression, i.e., the vertical spread of the individual perceived risks around the regression line (small SDs are desirable).
In multiple linear regression, the perceived social/internal consequences of refusing to share were associated with both RSS and RPS, after controlling for other psychosocial constructs and demographic variables.
In the multivariate ordinal logistic regression analysis, perceived health was no longer associated with pet ownership.
However, we find no collinearity between traumatic events and perceived needs in the regression analyses.
In both our correlation and regression analyses, perceived social capital consistently emerged as an indicator associated with HIV ART adherence.
To estimate associations between regular use of wireless phones, health symptoms and perceived health ordinal logistic regression adjusted for age and sex was used.
Correspondingly, when dichotomizing the Borg scale into "low or moderate" (Borg < 4) and "high" (Borg ≥ 4) perceived exertion, the logistic regression analysis demonstrated strong association between perceived exertion and high neck/shoulder muscle activity.
Using multiple logistic regression with perceived need for dental treatment as outcome variable, all OIDP items and family wealth index were entered simultaneously whilst controlling for age, gender, place of residence and parental education.
