Is there a good way in R to impute values which exist, but are less than the detection level for an assay? Thanks, Jonathan Williams OPTIMA Radcliffe Infirmary Woodstock Road OXFORD OX2 6HE Tel +1865 (2)24356
> -----Original Message----- > From: Jonathan Williams > [mailto:jonathan.williams at pharmacology.oxford.ac.uk] > Sent: 15 March 2004 12:48 > To: Ethz. Ch > Subject: [R] imputation of sub-threshold values > > > Security Warning: > If you are not sure an attachment is safe to open contact > Andy on x234. > There are 0 attachments with this message. > ________________________________________________________________ > > Is there a good way in R to impute values which exist, > but are less than the detection level for an assay? > Thanks, >There's no failsafe way to do this in ANY language. Imputing using the value of the detection level, or half way between there and zero (for strictly positive parameters) are two methods commonly used in the pharma industry, if the proportion of results below detection level is small. Simon Fear Senior Statistician Syne qua non Ltd Tel: +44 (0) 1379 644449 Fax: +44 (0) 1379 644445 email: Simon.Fear at synequanon.com web: http://www.synequanon.com Number of attachments included with this message: 0 This message (and any associated files) is confidential and\...{{dropped}}
On Mon, 15 Mar 2004, Jonathan Williams wrote:> Is there a good way in R to impute values which exist, > but are less than the detection level for an assay?If there were a good way to do it, it would probably be implementable or implemented in R. If you can persuade the people measuring the values to give you the numbers (assuming they are just below `limit of detection' rather than genuine non-detects) you will reduce the need for imputation. This is often the most powerful technique -- analytical chemists are trained not to give out numbers less than some multiple of the measurement error, but they can often be persuaded that statisticians can be trusted with these numbers. If you think your data are close to one of the parametric survival models in the survival package, you can analyse the data as left-censored, without imputation. You could impute from the fitted model, if you need imputations. Otherwise you may be stuck with some ad hoc imputation with a sensitivity analysis to see how the results depend on the imputed value. -thomas Thomas Lumley Assoc. Professor, Biostatistics tlumley at u.washington.edu University of Washington, Seattle