arXiv subjects

Kishore Kulat

Publications and source records attributed to Kishore Kulat.

2 recordsLinked to original sources

arXiv2017

PSF : Introduction to R Package for Pattern Sequence Based Forecasting Algorithm

This paper discusses about an R package that implements the Pattern Sequence based Forecasting (PSF) algorithm, which was developed for univariate time series forecasting. This algorithm has been successfully applied to many different fields. The PSF algorithm consists of two major parts: clustering and prediction. The clustering part includes selection of the optimum number of clusters. It labels time series data with reference to such clusters. The prediction part includes functions like optimum window size selection for specific patterns and prediction of future values with reference to past pattern sequences. The PSF package consists of various functions to implement the PSF algorithm. It also contains a function which automates all other functions to obtain optimized prediction results. The aim of this package is to promote the PSF algorithm and to ease its implementation with minimum efforts. This paper describes all the functions in the PSF package with their syntax. It also provides a simple example of usage. Finally, the usefulness of this package is discussed by comparing it to auto.arima and ets, well-known time series forecasting functions available on CRAN repository.

stat.ML↗

arXiv2016

R package imputeTestbench to compare imputations methods for univariate time series

This paper describes the R package imputeTestbench that provides a testbench for comparing imputation methods for missing data in univariate time series. The imputeTestbench package can be used to simulate the amount and type of missing data in a complete dataset and compare filled data using different imputation methods. The user has the option to simulate missing data by removing observations completely at random or in blocks of different sizes. Several default imputation methods are included with the package, including historical means, linear interpolation, and last observation carried forward. The testbench is not limited to the default functions and users can add or remove additional methods using a simple two-step process. The testbench compares the actual missing and imputed data for each method with different error metrics, including RMSE, MAE, and MAPE. Alternative error metrics can also be supplied by the user. The simplicity of use and significant reduction in time to compare imputation methods for missing data in univariate time series is a significant advantage of the package. This paper provides an overview of the core functions, including a demonstration with examples.

stat.ME↗