简体繁体 English

R中的预测有马

[英]forecasting arima in R

原文 2012-08-27 00:21:06 9 1 r/ time-series/ forecasting

I tried using the forecast package in R. When plotting the forecast for a number of samples in the future, I get a constant function. 我尝试在R中使用预测包。在将来绘制多个样本的预测图时，我得到一个常数函数。 What is the reason of that? 是什么原因呢？

1 个解决方案

It is a FAQ and so much so that Rob Hyndman (who is the author of the forecast package) felt compelled to write an entire blog post about it: flat forecasts . 这是一个FAQ，而且如此之多，以至于Rob Hyndman（天气预报包的作者）感到被迫撰写有关此内容的整个博客文章：平稳的预报。 Allow me to cite: 请允许我引用：

It is possible, even likely in some circumstances, that the future observations will have the same mean and then the forecast function is flat. 将来的观察结果可能甚至在某些情况下具有相同的均值，然后预测函数将保持不变。

A random walk model will return a flat forecast function (equal to the last observed value of the series). 随机游动模型将返回平坦的预测函数（等于该序列的最后观察值）。

An ETS(A,N,N) model will return a flat forecast function. ETS（A，N，N）模型将返回固定的预测函数。

An iid model will return a flat forecast function (equal to the mean of the observed data). 一个iid模型将返回一个固定的预测函数（等于观察到的数据的平均值）。

This is not a bug. 这不是错误。 It is telling you something about the time series — namely that there is no trend, no seasonality, and insufficient temporal dynamics to allow the future observations to have different conditional means. 它告诉您有关时间序列的信息-即没有趋势，没有季节性，并且时间动态不足以使将来的观测具有不同的条件手段。