Eviews时间序列分析实例（An example of Eviews time series analysis）.doc

Eviews时间序列分析实例（An example of Eviews time series analysis） Time series is a kind of data form that is often involved in the market forecast. The seventh chapter gives a more detailed introduction to it. Through the study of the seventh chapter, the reader knows what the time series is, and comes into contact with the principles of time series analysis and some examples. The main content of this section is how to use Eviews software for analysis. Example of exponential smoothing So called exponential smoothing is actually a weighted average of historical data. It can be used for any short-term prediction that has no obvious function rule but does have some kind of pre - and post - correlation time series. Since many other methods of analysis do not have this feature, exponential smoothing still plays an important role in time series forecasting. () - exponential smoothing Single exponential smoothing, also known as exponential smoothing. Its most outstanding advantage is that the method is very simple, and even if the smoothing value at the end of the sample is obtained, the prediction results can be obtained. One exponential smoothing is the ability to track changes in data. This characteristic has all indices. When the latest sample data is added to the prediction process, the new data should replace the old data, and the old data will gradually occupy a secondary position until it is eliminated. In this way, the prediction value always reflects the latest data structure. Primary exponential smoothing has limitations. First, the predictive value can not reflect the change of trend, seasonal fluctuation and regular changes; second, this method is more suitable for short-term prediction, and is not suitable for long-term forecasts; third, the predictive value is the mean of historical data, and therefore change the actual sequence compared with hysteresis. Whether the exponential smoothing is ideal depends largely on the smoothness factor. Eviews provides two ways to