An Epitomized Approach to Possess Promising Predictions by using Time-Series Analysis and Forecasting in R language

Author Name(s): Ramachandra Rao Kurada, Dr. Karteeka Pavan Kanadam
Author Email: ramachandrarao.kurada@gmail.com

Abstract

The aim of this work is to exertion a plug-in, formerly named as Time Series Analysis and Forecasting (TSAF) and incorporates this plug-in into R language. The intent behind materializing this plug-in is to establish a firstrated approach to forecast in-advance extrapolations in time series data and to make accurate decisions methodically. The plug-in provides a computationally intelligent environment by accepting a preprocessed time series datasets as input and sense the direction of outputs that will transpire over the coming ages. The internal code structure and implementation details in between the input and output precincts are factorized with the general machine learning, statistical calculation, and visualization packages. The preeminence of this incarnated viewpoint is scientifically verified over timeseries datasets archived in UCI repositories. The results enabled from these datasets pertain to revive qualitative nature of forecasting, which helps the users to predict or foresee changing domain trends and thereby make strategic decisions and hopefully gain lifelong encroachments in this process.

Keywords

Time series analysis, Forecasting’s, Predictions, Machine Learning, Statistical study, Computational Intelligence.

Introduction

A time series data is a sequence of numerical data points in successive order, measured at regular intervals, with no minimum or maximum amount of time [1]. Parenthetically, such data is assorted into one of the three sets as a) Time series: Holds a set of annotations on the data for an attribute to changes over time, b) Crosssectional: Data of more than one or more characteristic is composed at a similar point in time and c) Pooled: Hybrid data with the combination of both time series and cross-sectional data [2].

A dependability study to comprehend the factors of trend, seasonality, cyclic effect and noise in time series data inclining statistical technique is called time series analysis. In general, Trend is a pattern in time series data, which is linear or non-linear, indicates the change in data either ascending or descending. Seasonal is a significant pattern in time series data, always applicable to known period and is influenced by an annual factor of data and time. Seasonality is effected systematically in a fixed interval period. Cyclic pattern effects the time series data in a changing interval time as fluctuations concerning hours, day, weeks, months, quarters and years [3].

Conclusion

This paper deliberated a premeditated approach to make predictions and analysis over time series data with a TSAF plug-in. All the essential elements such as computational logic with machine intelligence primitives, statistically significant methods are incorporated into this plug-in. The user supplies the time series data as an input, and abstractly after computation of the internal logic without any intervention, the final outputs are generated in the form of time series patterns such as trend, seasonal, cyclic components. This plug-in not only figures the time series patterns but does optimized forecasting more precisely. The usage of this plug-in is thoroughly tested by imposing real-time datasets as inputs and clear-cut results, and predictions are exhibited in the previous section more invariably by revitalizing the unique abilities of each method. For novel investigates, this chaperone can be used as a platform to practice and demeanor research in this direction by having a quick insight into the projected methodology.

1,353 total views, no views today

Download PDF File

About the author: admin