A Pliable Lasso
A Pliable Lasso
复制标题
DOI：
10.1080/10618600.2019.1648271
复制
发表时间：
2019-09-03
影响因子：
2.4
通讯作者：
Friedman, Jerome
中科院分区：
文献类型：
--
作者：
Tibshirani, Robert;Friedman, Jerome
我们提出了一个概括的套索，允许模型系数作为一个函数的一般设置的一些预先指定的修改变量。这些修饰符可能是性别、年龄或时间等变量。该范例是相当通用的，每个套索系数由修改变量Z的稀疏线性函数修改。模型以分层方式进行估计，以控制自由度并避免过拟合。修改变量可以被观察到，仅在训练集中观察到，或者整体上未观察到。我们的建议与变系数模型和高维相互作用模型有联系。我们提出了一个计算效率高的算法，其优化，精确的筛选规则，以方便应用到大量的预测。该方法说明了一些不同的模拟和真实的例子。可以在网上找到。
We propose a generalization of the lasso that allows the model coefficients to vary as a function of a general set of some prespecified modifying variables. These modifiers might be variables such as gender, age, or time. The paradigm is quite general, with each lasso coefficient modified by a sparse linear function of the modifying variables Z. The model is estimated in a hierarchical fashion to control the degrees of freedom and avoid overfitting. The modifying variables may be observed, observed only in the training set, or unobserved overall. There are connections of our proposal to varying coefficient models and high-dimensional interaction models. We present a computationally efficient algorithm for its optimization, with exact screening rules to facilitate application to large numbers of predictors. The method is illustrated on a number of different simulated and real examples. for this article are available online.