Market Prediction model
- Concept of market prediction : Market prediction is overall estimation of competence or demand of products through competitors’ efforts or specific marketing tactics. Market prediction makes policies as company's marketing, new product development and future investment.
- The purpose of market prediction is to offer basis on which market personnel makes decision in the future. According to demand, which can be short-term or long- term or middle period too.
- Methods of market prediction include rolling average law , index level and smooth law , Del Philippines law and Regression Law and etc.
- The law of rolling average includes: simple rolling average, weighted moving average law, trend revising law of rolling average and rolling average law two times. Rolling average law dispels original accidental influence of factor to adopt, through sample smooth and prediction. This kind of method is a kind of commonly used prediction methods to short-term forecasting.
- weighted moving average
The law of rolling average of once (simple ) does not consider the recent market situation and different influences on the predicted value of the long-term market situation. In fact, the statistics in the near past can predict the variation tendency of the market accurately, and the statistical data in a long distance may influence the predicted value to a very low degree, so that those in a long distance period of time can neglected . Because of this the law of weighted moving average counts the recent samples for greater right , calculating the average of moving again on this basis, and the predicted value will be more close to the real fact.
The predicted value of law of weighted moving average can be calculated according to the following formula:
- Trend revising law of rolling average
It is on the basis of the law of rolling average and weighted moving average law of once (simple ) that the trend revising law of rolling average is produced, calculating the trend value of the change, asking out the averages of change trend of several issues, and then utilizing this average of trend to revise rolling average and predicted value of the weighted moving average of once (simple ), in order to dispel a kind of computing technology of original predicted value lagging influence.
The rolling average law two times is to carry on rolling average once again on a foundation of the averages. It can calculate according to the following formula :
- 指数平滑法
Smooth law of the index is based on historical data and information, predicting the future market with smooth coefficient.
The smooth law of the index includes smooth law of index once and smooth law of index two times.
The smooth law of index once can be calculated according to the following formula:
- Smooth law of index two times
Smooth law of index two times is the second smooth on the basis of the first time smooth process. We can get the these two values and set up the following model of predicting according to smooth coefficients :
