At the species level, there was an average of 58.7% missing data, as measured by number of base pairs, but only 49.9% when measured in terms of PI characters expected from the missing regions.
Subjects with missing data on measured anthropometric characteristics at enrolment and recalled weight at age 20, and women aged >80 years at baseline were also excluded.
With structured data, almost every data field can be analyzed, missing data can be measured, and the ratio of information to data is very high.
First, regression imputation methods were used to impute missing data on covariates measured at baseline, and all covariate adjustment models were computed with the missing data replaced by the imputed values.
We used the proportion of discrepant answers and missing data as measures of misclassification (unlike [ 6] who used hypothesised specificity and sensitivity).
Students with missing data on weighing/height measuring history or recall ability were coded missing in the combined variables.
Using all data (measured concentrations, missing data types I III, and covariates), we create the log-likelihood function 1, solve for the MLEs of β and σ (denoted β̂ and ς̂), and impute a value by randomly sampling from a log-normal distribution with the estimated parameters.
For categorical and continuous variables that had greater than 2% missing data and were not measured from one day to the next, a "never measured" category was created and risk was compared to the other categories or cut-points and pooled into the appropriate reference group.
We measured missing data in two ways: missing data per gene region and per species.
The dataset contains very few missing data (average 72% complete, measured as the percentage of positions with data present within the total alignment), especially in the case of the newly sequenced taxa (average 82% complete, all but two >68% complete).
