causality.html
来自「时间序列工具」· HTML 代码 · 共 30 行
HTML
30 行
<html><head><title>The causality window</title></head><body bgcolor=white><h1 align=center>The meaning of the causality window</h1>This parameter is relevant for programs which estimate the forecasterrors for a given model. Suppose one wants to forecast the elementn<sub>j</sub> of a time series. Since one implicitly supposes that thiselement is unknown, one should not use any information about thiselement to build the model. Therefore one should exclude all delayvectors containing this element from the model building process. Thisis realized by the routine <ahref="../../source_c/routines/exclude_interval.c">exclude_interval</a>.<br>This exclusion of parts of the vectors from the model building processcan be a severe problem if one doesn't do a one-step forecast, but an-step forecast, with n large. In this case one has to exclude a largepart of the data (roughly 2*n) and it could happen that the rest ofthe data is not sufficient for a reasonable model creation. By meansof the <font color=blue>-C</font> flag one can therefore shorten thiswindow. Setting the parameter to zero corresponds to a window of aone-step forecast.<br><center><font color=red>Change this parameter only if you really knowwhat you are doing!</font> </center></body></html>
⌨️ 快捷键说明
复制代码Ctrl + C
搜索代码Ctrl + F
全屏模式F11
增大字号Ctrl + =
减小字号Ctrl + -
显示快捷键?