Class | Analysis and data manipulation command |
Name | impute |
Arguments | <ytrait> |
<ytrait> on <loc1>...[to]...<locN> [complete_obs] |
If a single argument is given, replace missing quantitative trait values with the predicted values from a regression on the spouse, sibling and offspring observed values. Designed mainly for imputing missing age or date of birth.
If covariates are given, then replace missing quantitative trait values with the predicted value from the multivariate regression on the list of predictors (which may include the average allele length of a marker locus). The com option means only individuals with no missing values for any of the listed predictor traits will be updated. Otherwise, missing values are replaced with the sample mean for that phenotype when calculating the predicted value.
Example:
>> impute DOB Relation N Intercept Sex Beta Offspr 22780 -10631.6323 700.7983 0.9357 Siblings 44283 14.0954 -26.4170 0.9999 Spouses 18893 -1161.2528 1456.5714 0.9617 Outliers (>3 SDs) Regression ID Observed Expected StdRes Relatives-Mean ---------- ---------------------- -------------- -------------- ------- -------------- Offspring 0002.00.009-00.009.013 7305.000000 -28005.154070 3.67 -18567.000000 Offspring 0002.01.004-01.004.012 7305.000000 -28403.503637 3.71 -18992.714286 Spouse 0004.02.004-02.004.017 -23741.000000 10832.702214 -3.59 10957.000000 Imputed 9673 missing values.
>> include williamsex.in >> impute adjChol = ldl CHD age ------------------------------------------------ Linear regression analysis of trait "adjChol" ------------------------------------------------ Variable Beta Stand Error t-Value ----------------------------------------------------- Intercept 239.4865 44.2594 5.4110 *** ldl*2 120.8932 36.1504 3.3442 ** CHD 27.0280 117.2997 0.2304 . age -3.9219 1.6702 2.3481 * No. usable observations = 27 ( 45.0%) Model Mean Square = 49742.0342 (df= 3) Mean Square Error = 12360.3563 (df= 23) Multiple R**2 = 0.3442 Akaike Inf. Criterion = 9.6445 Wrote 14 predicted values to adjChol.
<< (predict) | Up to index | >> (kaplan-meier) |