×
Suggest this Article to:
TOP INDEXERS
Public Article
-
verified
SEASONAL ARIMA MODEL FOR FORECASTING MONTHLY TEMPERATURE FOR COIMBATORE, TAMIL NADU
ISSN: 2277 - 7601
Publisher: author
SEASONAL ARIMA MODEL FOR FORECASTING MONTHLY TEMPERATURE FOR COIMBATORE, TAMIL NADU
Indexed in
Biology and Life Sciences
ARTICLE-FACTOR
1.3
Article Basics Score: 3
Article Transparency Score: 3
Article Operation Score: 2
Article Articles Score: 3
Article Accessibility Score: 3
SUBMIT PAPER ASK QUESTION
International Category Code (ICC):
ICC-0402
Publisher: Kiran Abasaheb More
International Journal Address (IAA):
IAA.ZONE/2277392217601
eISSN
:
2277 - 7601
VALID
ISSN Validator
Abstract
Seasoanl Autoregressive integrated moving average (SARIMA) is a data mining technique that is generally used for time series analysis and future forecasting. In this paper, the SARIMA based weather forecasting tool has been developed by implementing the SARIMA algorithm in R. An attempt has been made to develop a SARIMA model to long term temperature data of Coimbatore for the period of twenty eight years (1991-2018). As the SARIMA model works only on stationary data, the data should be trend and seasonality free.The stationarity test were tested using ADF ,KPSS and PP test and found to be stationary at six and five lag. The analysis revealed that the best seasonal models which are satisfactory to describe the data are SARIMA(3,0,3)(1,1,1)12for monthly maximum and SARIMA(2,0,2)(1,1,2)12 for monthly minimum temperature data respectively.The root means square error values were 0.8 and 0.6 for temperature data which showed that the algor...
