Analysis
Because of the sampling frame, the data must be
weighted when generating descriptive statistics. In the
multivariate analysis, special techniques must be
employed to account for the five implicate data sets
created by the multiple imputation procedure used by the
Federal Reserve. The multiple imputations are
repetitions drawn to simulate a Bayesian distribution of
the missing values under a model. Therefore,
appropriately combining analyses of each data set
completed by imputation yields an approximately valid
Bayesian inference under that model (Rubin, 1987). This
repeated-imputation inferences (RII) technique provides
“averaging” rules for the analysis. More specifically, the
multiple imputed values are averaged to produce the best
estimate of what the results would have been if the
missing data had been observed, and the variance
estimates are corrected for the uncertainty due to missing
values. Montalto and Sung (1996) provide a more
detailed discussion of practical applications of RII in theTwo separate OLS analyses were run to estimate the
APR and the potential dollar savings for revolvers.
Separate OLS analyses were run for each of the five
implicates. Then, using the RII technique, estimates
were derived from all five implicates, and the variability
in the data due to missing values and imputation was
incorporated in the estimation. SAS/IML programming
was utilized for implementing the RII.
Analysis
Because of the sampling frame, the data must be
weighted when generating descriptive statistics. In the
multivariate analysis, special techniques must be
employed to account for the five implicate data sets
created by the multiple imputation procedure used by the
Federal Reserve. The multiple imputations are
repetitions drawn to simulate a Bayesian distribution of
the missing values under a model. Therefore,
appropriately combining analyses of each data set
completed by imputation yields an approximately valid
Bayesian inference under that model (Rubin, 1987). This
repeated-imputation inferences (RII) technique provides
“averaging” rules for the analysis. More specifically, the
multiple imputed values are averaged to produce the best
estimate of what the results would have been if the
missing data had been observed, and the variance
estimates are corrected for the uncertainty due to missing
values. Montalto and Sung (1996) provide a more
detailed discussion of practical applications of RII in theTwo separate OLS analyses were run to estimate the
APR and the potential dollar savings for revolvers.
Separate OLS analyses were run for each of the five
implicates. Then, using the RII technique, estimates
were derived from all five implicates, and the variability
in the data due to missing values and imputation was
incorporated in the estimation. SAS/IML programming
was utilized for implementing the RII.
การแปล กรุณารอสักครู่..
