Performance comparison of TSA methods using past weather data (1970–2015) over the manaus region of Brazil

Over the centuries, due to the use of fossil fuels that release carbon dioxide and other greenhouse gases, climate is rapidly changing. Due to the climate change, most weather parameters experience variation. For human beings, it is necessary to have the knowledge about precise weather forecasts for planning their daily activities. While the intense behaviour of weather is challenging for prediction, temperature, wind speed and humidity are three major weather parameters that influences in manifold manner. The proposed paper targets a comparative of various time series methods. The model is developed on one dataset, to identify the best suitable time series method for prediction. Manaus database is collected from the year 1970 to 2015 to check the efficiency of time series methods. Error measures such as Normalized Mean Square Error (NMSE) and Normalized Root Mean Square Error (NRMSE) are calculated to provide an insight into the prediction values.