Forecasting

Forecasting is an attempt to get an idea of ​​what will happen in the future. In this case the image obtained will be a reference to making a decision. In conditions of uncertainty is difficult for us to determine an effective planning. Forecasting can help leaders to reduce uncertainty in planning.
Forecasting is a process to estimate how future needs including the needs of the measure of the quantity, quality, time and location that is required in order to meet the demand for goods or services. In the business world, forecasting is the basis for capacity planning, budgeting, sales planning, production and inventory planning, resource planning, planning the purchase or procurement of raw materials, and so on.
Forecasting function to make predictions needs (demand) of the product to be made which are expressed in quantity (amount) of product as a function of time. Forecasting carried out in the long term (long term), medium term (medium term), and short-term (short term). Estimates relating to the statement (1) what Will be demanded, (2) how many, and (3) Pls supplied it Should Be? Monitoring forecasting is needed by doing a comparison between the predicted requirements with the actual. For that act of correction can be done immediately to the needs of the forecast.
Forecasting methods can be classified into two categories, namely Qualitative and Quantitative. This qualitative method is a method to make forecasts, but without historical data (previous sales periods). For example we have just set up the company. We do not know how much will we sell for the next period? we can predict by conducting market research or the Delphi method.
As for the quantitative, we can use data data historical period (the period prior to forecasting). Of course there are models and mathematical calculations to predict the forecast results. Can use regression models, Econometric, Time series analysis. In addition to the above two categories of methods we can perform simulation modeling predictions forecast to know the results.
The first and main concepts of the forecaster (astrologer) is: Forecasting is always wrong, but some results can be useful forecasting. Forecasting done because we have to make a decision well before the actual information that is needed there. If we wait until the information is available most likely already too late.
Things to do in the forecast is to collect data of the past and develop a forecasting model with more than one technique. Then select the model validation test and the most accurate and who elected to use the model to forecast the actual reply. The new data will actually appear after a few periods later
Forecasting accuracy measurements are usually performed to compare and choose among two or more forecasting techniques. If we only had one technique at hand, the measurement accuracy of forecasting does not have much meaning because we would not choose anything.
The concept is basically, forecasting models are considered accurate if the error peramalannya small, or more precisely, if we have two forecasting techniques, A is considered more accurate than B if A technique forecast error is smaller. Forecasting error is calculated from the difference between the forecast and actual data.
The most popular method is to look at patterns of data. Who have applied techniques include linear regression, nonlinear regression, Bayesian Regression, ARIMA (Box Jenkins), Moving Average, etc..
One of the main assumptions in the use of the techniques above are that: the data pattern that occurred in the past will be repeated in the future. In other words, if the assumptions are not met then the most likely forecast results become inaccurate.
Unfortunately there is no best technique for all cases of forecasting.
From the technical side, good is accurate. Many studies show that more sophisticated techniques are not necessarily more accurate. The way the ‘most correct’ to measure the forecasting accuracy by comparing the results of course are the forecasts with actual conditions in the field.