Table 4:
Overview of available software for performing regression calibration
| Package/ procedure | Website location or information | Notes | References |
|---|---|---|---|
| Procedure rcal within the STATA package merror |
http://www.stata.com/merror/ See also: http://www.stat.tamu.edu/~carroll/eiv.SecondEdition/statacode.php |
For generalized linear models with X* having classical measurement error and (a) the error variance is known; (b) replicate measurements of X* are available or (c) when there is also available an ‘instrumental variable’ that is correlated with X and whose errors are independent of the errors in X*. | Hardin et al (2003)117 |
| Procedure eivreg within STATA | http://www.stata.com/manuals13/reivreg.pdf | For linear regression where X* has classical measurement error and the error variance (or ratio of the error variance to total variance) is known. | Hardin et al (2003)117 |
| NCI SAS macros |
https://epi.grants.cancer.gov/diet/usualintakes/macros.html |
(a) For X* measured in all individuals that, after suitable transformation, satisfies a linear measurement error model, together with an X** measured in a subsample that, after suitable transformation, has classical measurement error. (b) For X* measured in all individuals that, after suitable transformation, has classical measurement error. A substantial subsample should have at least one repeat value of X*. In these options X* and X** may be univariate, bivariate or multivariate. X* and X** may have excess zeros. |
Kipnis et al (2009)22 |
| Spiegelman SAS macro %blinplus |
https://www.hsph.harvard.edu/donna-spiegelman/software/ | For univariate or multivariate X* measured in all individuals in the main study and a validation study where both X and X* are measured in all individuals. X* satisfies the linear measurement error model (2). | Rosner et al (1990)93 |
| Spiegelman SAS macro %relibpls8 |
https://www.hsph.harvard.edu/donna-spiegelman/software/ | For univariate or multivariate X* measured in all individuals and repeat measurements of X*. X* satisfies the classical measurement error model (1). | Rosner et al (1992)94 |
| Spiegelman SAS macro %rrc |
https://www.hsph.harvard.edu/donna-spiegelman/software/ | For a time-varying covariate X in a Cox regression model. X* satisfies the linear measurement error model (2). The method uses risk-set regression calibration for estimating the risk parameters (see Section 5.1.4). | Liao et al (2011)108 |