I am a beginner to model fitting, and I have been working on generating a model for a CO2 emissions data set. The distribution of the data points in a number of these columns are very markedly dual-peaked. One of the columns with a dual peak

There is an obvious geometric reason (in the context of this particular dataset and solar azimuth angles) that explains these peaks, but I wanted to know the best general strategy, as well as code to use (in python) to take when imputing data into columns such as this one with many peaks. Obviously using the mean or most common value will skew the dataset in a way that isn't consistent with the observable behavior of the data.

  • $\begingroup$ Maybe hot deck imputation. $\endgroup$ Aug 2, 2023 at 13:41


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.