Fig 1.
Example Activity Patterns with Same Means and Different Variability Patterns.
The solid line represents example pattern 1) a retired person who does housework on most days, but who never engages in structured exercise. The dashed line represents example pattern 2) an older person who spends most of the day using the computer, but has bouts of vigorous outdoor gardening and home repairs when the weather allows.
Fig 2.
A Visual Depiction of the Mixed-Effects Location-Scale Model.
Effects on the left side are for the location model; effects on the right side are for the scale model for the residual variance. Curly brackets represent differences; square brackets represent (approximate) residual variance. Dashed lines represent participant-specific effects; solid lines represent sample average (or fixed) effects. Dots represents participant-specific observed physical activity; squares represent the average of the dots. γ00 = location-model fixed intercept. β0,i = participant-specific location-model intercept. e0,i = participant-specific residual value. u0,i = participant-specific location-model random intercept value (β0,i−γ00). δ00 = fixed intercept of the residual variance. σ2e,i = participant-specific residual variance. ω0,i = participant-specific scale-model random intercept value (σ2e,i−δ00).
Table 1.
Group-Specific Descriptive Statistics for Participants included in the Final Model.
Table 2.
Results of the Final Mixed-Effects Location-Scale Model.
Fig 3.
Between-Sex and Between-AD Status Differences in the Variability of Physical Activity Across Occasions.
Man and woman with mild AD (top) and a healthy man and woman (bottom).