Sample size calculations based on an alpha of 0.05, 10 predictor variables, expected effect size of 0.20 (moderate effect size), and power of 0.80 indicated a required sample size of 91 subjects.
One way of improving estimates beyond the 250 Mg ha−1 range is by including lidar measurements with higher resolution such as those from GEDI (expected launch in 2018) [32] or ICESAT-2 (expected launch in 2017) as predictor variables individually or through fusion with other datasets.
One way of improving estimates beyond the 250 Mg ha−1 range is by including lidar measurements with higher resolution such as those from GEDI (expected launch in 2018) [ 32] or ICESAT-2 (expected launch in 2017) as predictor variables individually or through fusion with other datasets.
As expected, the RRs for other predictor variables examined did not change between this study and our prior analysis.
As expected, as the number of predictor variables decreases, the aROC and sensitivity and specificity results worsen.
Statistically significant paths were expected to proceed from the exogenous predictor variables (AIS, COG, and MOT) to the potentially mediating participation variables (mobility, occupation, and social integration) to the endogenous outcome variables (LSI, HEALTH).
With a significant large influence of 4-5 predictor variables (as is expected in this study) and a squared multiple correlation coefficient between 0.5 and 0.7 between 25 and 65 patients should be included.
The construct validity of the measures will be assessed prior to analysis by examining correlations between predictor variables that are expected to be similar (convergent validity) and dissimilar (discriminant validity).
It is expected that as models increase in complexity by adding predictor variables, a significant decrease in Deviance is expected to occur, and the significance of the decrease is tested with a chi-square test [ 35].
This method greatly improves performance in the case of highly correlated predictor variables (as we expect to have in clinical data), through the identification of groups of phenotypes with significantly high correlation that contribute the most to the variation in the data.
