Subsequently, markers with 12 or more missing data points (approximately 10% or more missing genotypes) were also removed from the analysis, as the maximum likelihood mapping algorithm initially employed for map construction may be sensitive to having many unknown genotypes in the dataset [ 65].
Another challenge was the missing data on approximately 30% of tissue cores from the tissue microarrays.
For the eight-taxon data set with missing data included, approximately 58% of the sites support position B (the ML tree for this data set).
With sites with missing data excluded, approximately 78% of the remaining sites support position S, the ML tree in this case.
To reduce potential bias associated with missing data, missing values (approximately 0.1 40%) were multiple-imputed (n=5 imputed datasets).
When all covariates were controlled simultaneously, the magnitude of the results did not change (e.g., RR = 1.12 for alcohol only, RR = 0.80 for marijuana only, and RR = 1.24 for alcohol plus marijuana), though confidence intervals were wider due to approximately 19% missing data.
The proportion of missing data across the matrix is approximately 70%%.
The overall data set has approximately 20% missing data, but gives unbiased estimates of parameters based on Migrate-N 3.2.6.
The average interval between the first and second was 55 days (range: 40 – 100 days) and missing data across assessments was low, approximately 2%.
26 As expected in this kind of registration, which takes place in a busy clinical setting, missing data were a weakness with approximately 28% of the patients in the DANBIO registry excluded from the study population.
