Time-Series analysis for wind speed forecasting

Print   Print  

Authors :

Garima Jain1

Author Address :

1Department of Computer Science and Engineering, Swami VivekanadSubharti University, Meerut-250005, India.

Abstract :

In this paper, an enormous amount of study has been made on various weather forecasting models and many specialists developed different models for optimal results. Different models which is taken for implementation and were used for predicting the forecasted data. A technique used for forecasting the given data is defined as a time series data. Box-Jenkins method is a statistical methodology used for prediction of data in time series. During this paper, an ARIMA model is implemented for predicting the data of Wind Speed. Results are compared with respective model i.e.., ETS Model. With this paper we like to through some light on ARIMA (Auto-Regressive Integrated Moving Average) and ETS (Exponential Smoothing) models for forecasting the weather conditions in India. ARIMA model is chosen; because of it is acceptable in terms of easiness and extensive of the model. Eight years weather data (from year 2007 to 2014), i.e., wind Speed for time-intervals to forecast (i.e.., 1 hours) are used in this research.

Keywords :

ARIMA (Autoregressive Integrated Moving Average), ETS (Exponential Smoothing), AIC (Akaike’s Information Criteria), BIC (Bayesian Information Criteria), RMSE (Roo

DOI :

10.26637/MJM0S01/11

Article Info :

Received : December 24, 2017; Accepted : January 21, 2018.