NAs in x and y coordinates

[tmalsburg]

Example:

data(samples)

fixations <- detect.fixations(samples)
head(fixations)

samples$x[100] <- NA
samples$y[100] <- NA
fixations <- detect.fixations(samples)
head(fixations)

What's the proper way to deal with NAs? We can't just delete samples with NAs because the velocity calculations assume that Δt between samples is constant.

[vasishth]

We can impute missing values given available data; I have done this in the
past.

Shravan Vasishth
Professor for Psycholinguistics and Neurolinguistics
Department of Linguistics
University of Potsdam, Germany
http://www.ling.uni-potsdam.de/~vasishth

On Mon, Mar 16, 2015 at 11:29 PM, Titus von der Malsburg <
notifications@github.com> wrote:

Example:

data(samples)

fixations <- detect.fixations(samples)
head(fixations)

samples$x[100] <- NA
samples$y[100] <- NA
fixations <- detect.fixations(samples)
head(fixations)

What's the proper way to deal with NAs? We can't just delete samples with
NAs because the velocity calculations assume that Δt between samples is
constant.

—
Reply to this email directly or view it on GitHub
#11.

[tmalsburg]

The problem is that there are different kinds of NAs that need different treatment but we don't know which type we're dealing with. For example, we'd like to keep NAs due to blinks, but whether blinks have NAs or not depends on the eyetracking system that was used.