Version 1.7 of the R package „imputeTS: Time Series Missing Value Imputation“ has been recently released on CRAN.
The package, which is developed within the SPOTSeven research group, provides several functions that help with missing data in time series. This includes simple and fast imputation algorithms, several advanced imputation algorithms and missing data visualization functions.
The imputeTS package can be found on CRAN. For installation execute install.packages(“imputeTS”) in R. A short introduction into usage can be found in the README.
User visible changes in version 1.7:
- Improved compatibility for input of advanced time series objects like zoo and xts
- Enabled multivariate input (data.frame, mts, matrix,…)
- Improved x-labels for na.distributionBar plot
- Revised user manual
- Revised usage examples
- Better worded error messages
For a complete list of all changes take a look into the package NEWS on CRAN.
Anyone who clicks on the link until August 24, 2016, will be taken to the final version of the article “Multi-fidelity modeling and optimization of biogas plants” (Martin Zaefferer, Daniel Gaida, Thomas Bartz-Beielstein) on ScienceDirect for free. No sign up or registration is needed – just click and read! Here is the link: http://authors.elsevier.com/a/1TK9q5aecSRzFl
Highlights of this article read as follows:
- Accurate and fast simulation models mandatory for the optimization of biogas plants.
- Improve precision of simulation models without increasing the number of evaluations.
- Combining results: complex simulator, simple estimation-based model and surrogate model.
- Advantages and limitations of multi-fidelity modeling approaches are discussed.