introduced by Engle(1982) and Bollerslev(1986), nowadays, models from the G(ARCH) class are the most popular volatility models among practitioners. GARCH models enjoy such popularity because they are capable of describing not only the feature of volatility clustering, but also certain other characteristics of financial time series, such as their pronounced excess kurtosis or fat-tailedness. researcheres don't have the talent as god to give any assumptions on the data which is generated from reality but they give assumptions when they are specifying models to replicate those statistical properties in the data. that's how the G(ARCH) class come from[see Chapter 3,Tsay's Analysis of financial time series for details in G(ARCH) class models].

我记得很久以前看到过资料区有一个eviews的教程，里面有告诉怎么用arch的，讲的很好，你可以去查一下

其实有的模型预报效果并不比ARCH类差，只是因为技术性的问题没有得到推广。

我对G（ARCH）了解的不多，不过倒是可以借此多讨论些