Forecasting and Analysis of Various Autoregressive Time Series Models

Abstract

Time series forecasting has a long track record in many application areas. In this thesis a particular interest is given to analyze and forecast various fields of data such as agricultural, metrological, renewable energy and medical data of Punjab, India. It also compares forecast performance of linear and non linear time series models. Punjab a state situated at north-western of India is very highly rich in agriculture and is the backbone of Indian economy. Its agricultural development is also due to the contribution of metrological condition of this state. But apart from its growth and competition in agricultural field, it also results in increase of cancer patients as well as other medical problems due to the excessive use of pesticides, disposal of industrial and agricultural waste. In this study all the aspects are taken into consideration the present study consists of six chapters according to various models and on its analysis bases. Chapter 1 is dealt with the Introduction part which includes general introduction, definitions of various useful terms and models, methodology, literature review and study area of the thesis. A brief description of the other chapters is as follows. newline Chapter 2 is spitted into two parts. In part A, statistical time series modelling techniques like moving average and least square method are used to study the future requirement of paddy and wheat. In part B, ARIMA model is used to forecasting average retail price of milk. In both the section, performances are evaluated in terms of Mean absolute error (MAE), Mean square error (MSE), Mean absolute percentage error (MAPE) and Root mean square error (RMSE). newline Chapter 3 is also spitted into two parts. In part A, the seasonal and periodic time series models are considered for statistical analysis of rainfall and temperature data of Punjab, India. At first, importance is given to forecast the future trend of maximum and minimum temperature of Punjab using non-seasonal