Forecasting Models for Yield Estimation

Only available on StudyMode
  • Topic: Forecasting, Time series, Fuzzy logic
  • Pages : 6 (1962 words )
  • Download(s) : 43
  • Published : January 12, 2013
Open Document
Text Preview
Prediction or forecasting is a common phenomenon for which all human beings are always eager to know. The pre-knowledge about unknown and uncertain future prepare them to cope up in an efficient way. Since the dawn of civilization, this desire has been satisfied by priests, astrologers, fortune tellers, etc. In the present scenario, the necessity of predicting future is fulfilled in ample ways. There are several forecasting methods available from simplest to some of the most complicated; from judgmental to quantitative. Forecasting, in true essence, is a branch of the anticipatory sciences used for identifying and projecting alternative possible future. It plays vital role in most of our activities and in all we do concerning the future. Weather prediction, staff scheduling, business, production planning and multistage management decision analysis are among distinctive examples of forecasting areas. In such fields people want to foresee as closely as possible and plan for the future. In broad terms, a forecast is simply a statement, based upon some criteria, concerning the future condition of a system. It opens menu windows onto future. It is a medium guiding towards plans for the development of a better future as the forecasted visions give an alternative to plan, design, shape, and cope with future. To make a forecast with 100% accuracy may not be possible, but efforts are made to reduce the forecasting errors or increase the speed of the forecasting process. For these forecasts, to be accurate, either no major change should occur from conditions that have prevailed during the past or such changes must be canceled out. Otherwise, forecasting errors are possible, unless some appropriate prudence about the direction of the forthcoming changes is developed. In the fast-paced and rapidly changing world, the future will be vastly different from the present in a number of ways. Furthermore, because of constant development of knowledge and advances in various arenas, the global society demands an increasing ability to shape the future for better. As a result, society and each institution recognize the need of knowledge about possible future which is basically the consequences of decisions and actions taken in present. Thus, it is increasingly necessary that one should have better forecasting tools that can be applied in an efficient way. It is more and more important to forecast the possible future, implied by the changes created by this knowledge. Hence, forecasting has become an essential tool for society to decide, plan, design, steer, manage, implement, and control changes by identifying preferable future with forecasts. In past few decades of research and development many methodologies and tools have emerged to deal with the forecasting processes. These methods can be broadly categorized as: a.Time series methods: Time series methods use historical data as the basis of estimating future outcomes. Such methods try to estimate how the sequence of observation will continue into the future. b.Causal / econometric methods: Some forecasting methods use the assumption that it is possible to identify the underlying factors which influence the variable that is being forecast. For example, sales of umbrellas might be associated with weather conditions. If the causes are understood, projections of the influencing variables can be made and used in the forecast. c.Judgmental methods: Judgmental forecasting methods incorporate intuitive judgments, opinions and subjective probability estimates. Some other methods like Simulation, Prediction market, Probabilistic forecasting and Ensemble forecasting, Reference class forecasting are also being employed for forecasting purposes. The time series analysis has been emerged as one of the useful tool to predict the future behavior of various systems. 1.1 :Time series analysis

A time series is basically, a collection of observations made sequentially in time. It is a chronological...
tracking img