Figure 2
Bias in the estimated inverse probability weighted log(hazard ratio) using the standard approach, multiple imputation for measurement error (MIME), and reparameterized imputation for measurement error (RIME) in settings where external validation data similar to external validation data set 2 are available, but the exposure prevalence in the validation data differs from the exposure prevalence in the main study (shown by vertical gray dashed line).

Bias in the estimated inverse probability weighted log(hazard ratio) using the standard approach, multiple imputation for measurement error (MIME), and reparameterized imputation for measurement error (RIME) in settings where external validation data similar to external validation data set 2 are available, but the exposure prevalence in the validation data differs from the exposure prevalence in the main study (shown by vertical gray dashed line).

Close
This Feature Is Available To Subscribers Only

Sign In or Create an Account

Close

This PDF is available to Subscribers Only

View Article Abstract & Purchase Options

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

Close