Quantitative forecasting is a tool use for future forecasting. The previous demand is a tool use to predict the estimate demand base on information. Types of Forecasting Methods Qualitative prediction quantitative forecasts depend on individual opinions of one or more experts
Quantitative Forecasting can be apply when there are the following three conditions:
Time series model is base on a series of sequential data is the same (e.g. weekly, monthly, yearly, etc.). This set of data which is a series of observations of a variety of variable according to time, usually tabulate and describe in the form of a graph that shows the behavior of the subject. Very precise time series use to forecast demand in the past request pattern quite consistent in a long period of time, so that the pattern will still be still continue.
Time series Analysis is base on the assumption that time series consist of component parts, namely: the pattern tendency (T), the pattern of cycle/cycle (C), the pattern of the season (S), the random Variation (R)
The model casual model is forecasting model which considering the variables that can affect the amount that is being predict. Or more simply that this method using the causal approach, and aims to predict conditions in future by finding and measuring some free variables (independent) are important, along with its effects on variables not free which will be foreseen.
Designing methods of forecasting are:
Once demand and sales are understood to be dynamic and not static, different methods of forecasting are needed. One such method is the moving average.
The moving average is use as a way to smooth forecasts into a more likely model. Consider the PDA example from the previous post. In that post we discussed how sales of PDAs from last year have dropped by 40%. (Read article HERE). Well, with the use of a moving average, the forecaster sales for the coming time period will not be lower by the same amount as the previous periods loss in sales
- Define the purpose of divination.
- scatter diagram Creation.
- Select at least two methods of forecasting that is consider appropriate.
- Calculate the parameters – parameters forecasting function.
- Calculate the error of each method of forecasting.
- Select the best method, that is, that have the smallest mistakes.
- Verify the forecasting.
Quantitative Forecasting can be apply when there are the following three conditions:
- a) the availability of information about the past
- b) that information can be quantify in the form of numerical data
- c) can be It is assum that some aspects of the pattern of the past will continue in the future.
Quantitative methods
Time series model
Time series model is base on a series of sequential data is the same (e.g. weekly, monthly, yearly, etc.). This set of data which is a series of observations of a variety of variable according to time, usually tabulate and describe in the form of a graph that shows the behavior of the subject. Very precise time series use to forecast demand in the past request pattern quite consistent in a long period of time, so that the pattern will still be still continue.
Time series Analysis is base on the assumption that time series consist of component parts, namely: the pattern tendency (T), the pattern of cycle/cycle (C), the pattern of the season (S), the random Variation (R)
Casual model (causal model)
The model casual model is forecasting model which considering the variables that can affect the amount that is being predict. Or more simply that this method using the causal approach, and aims to predict conditions in future by finding and measuring some free variables (independent) are important, along with its effects on variables not free which will be foreseen.
Causal methods there are three groups of methods that are commonly use:
- The method of regression and correlation wear technique of quadratic smallest (least square). This method is often use for short-term prediction. For example: predicting relationship amount credit given by current account, deposits and savings society.
- Econometric Methods base on the regression equations are approach simultaneously. This method is often used for the planning of the national economy in the short term and long term. For example: predicting the magnitude of monetary indicators create some years into the future, this is often done BI-party each year.
- input output Method use for long-term national economic planning. For example: predicting economic growth such as the gross domestic growth (GDP) for some period in the future 5-10 the coming year. Stages of design of forecasting in summary there are three stages that must be traverse in designing a method of forecasting.
Designing methods of forecasting are:
- The analysis on the data of the past. This step aims to get an idea of the pattern of the data concerned.
- Choose the methods to be used. There are various methods available with his hand. Different methods will produce a different prediction system for the same data. In general, it can be said that the successful method is a method which produces deviation (error) as small as small a between results prediction to reality that is happening.
- The process of data transformation from the past by using the selected methods. If necessary, appropriate changes held his needs.
Conclusion
Once demand and sales are understood to be dynamic and not static, different methods of forecasting are needed. One such method is the moving average.
The moving average is use as a way to smooth forecasts into a more likely model. Consider the PDA example from the previous post. In that post we discussed how sales of PDAs from last year have dropped by 40%. (Read article HERE). Well, with the use of a moving average, the forecaster sales for the coming time period will not be lower by the same amount as the previous periods loss in sales
No comments:
Post a Comment