線性工具變量的嶺徑估計(jì)

是 否 +2 論壇幣

k人 參與回答

經(jīng)管之家送您一份

應(yīng)屆畢業(yè)生專屬福利!

求職就業(yè)群

趙安豆老師微信：zhaoandou666

經(jīng)管之家聯(lián)合CDA

送您一個(gè)全額獎(jiǎng)學(xué)金名額~ !

立即領(lǐng)取

感謝您參與論壇問題回答

經(jīng)管之家送您兩個(gè)論壇幣！

+2 論壇幣

摘要翻譯：
本文給出了線性工具變量(IV)估計(jì)的漸近性態(tài)。通過將觀測數(shù)據(jù)分解為訓(xùn)練樣本和測試樣本，經(jīng)驗(yàn)地選擇正則化調(diào)整參數(shù)。在調(diào)整參數(shù)的條件下，訓(xùn)練樣本創(chuàng)建從IV估計(jì)器到先驗(yàn)估計(jì)器的路徑。最佳調(diào)諧參數(shù)是沿著該路徑使測試樣本的IV目標(biāo)函數(shù)最小化的值。經(jīng)驗(yàn)選擇的正則化調(diào)諧參數(shù)成為與感興趣參數(shù)聯(lián)合收斂的估計(jì)參數(shù)。調(diào)諧參數(shù)的漸近分布是非標(biāo)準(zhǔn)混合分布。Monte Carlo仿真表明，漸近分布捕捉到了抽樣分布的特征，當(dāng)這種嶺估計(jì)比兩級最小二乘估計(jì)性能更好時(shí)。
---
英文標(biāo)題：
《The Ridge Path Estimator for Linear Instrumental Variables》
---
作者：
Nandana Sengupta and Fallaw Sowell
---
最新提交年份：
2019
---
分類信息：

一級分類：Economics        經(jīng)濟(jì)學(xué)
二級分類：Econometrics        計(jì)量經(jīng)濟(jì)學(xué)
分類描述：Econometric Theory, Micro-Econometrics, Macro-Econometrics, Empirical Content of Economic Relations discovered via New Methods, Methodological Aspects of the Application of Statistical Inference to Economic Data.
計(jì)量經(jīng)濟(jì)學(xué)理論，微觀計(jì)量經(jīng)濟(jì)學(xué)，宏觀計(jì)量經(jīng)濟(jì)學(xué)，通過新方法發(fā)現(xiàn)的經(jīng)濟(jì)關(guān)系的實(shí)證內(nèi)容，統(tǒng)計(jì)推論應(yīng)用于經(jīng)濟(jì)數(shù)據(jù)的方法論方面。
--
一級分類：Statistics        統(tǒng)計(jì)學(xué)
二級分類：Machine Learning        機(jī)器學(xué)習(xí)
分類描述：Covers machine learning papers (supervised, unsupervised, semi-supervised learning, graphical models, reinforcement learning, bandits, high dimensional inference, etc.) with a statistical or theoretical grounding
覆蓋機(jī)器學(xué)習(xí)論文（監(jiān)督，無監(jiān)督，半監(jiān)督學(xué)習(xí)，圖形模型，強(qiáng)化學(xué)習(xí)，強(qiáng)盜，高維推理等）與統(tǒng)計(jì)或理論基礎(chǔ)
--

---
英文摘要：
  This paper presents the asymptotic behavior of a linear instrumental variables (IV) estimator that uses a ridge regression penalty. The regularization tuning parameter is selected empirically by splitting the observed data into training and test samples. Conditional on the tuning parameter, the training sample creates a path from the IV estimator to a prior. The optimal tuning parameter is the value along this path that minimizes the IV objective function for the test sample.   The empirically selected regularization tuning parameter becomes an estimated parameter that jointly converges with the parameters of interest. The asymptotic distribution of the tuning parameter is a nonstandard mixture distribution. Monte Carlo simulations show the asymptotic distribution captures the characteristics of the sampling distributions and when this ridge estimator performs better than two-stage least squares.
---
PDF鏈接：
https://arxiv.org/pdf/1908.09237

掃碼加我 拉你入群

請注明：姓名-公司-職位

以便審核進(jìn)群資格，未注明則拒絕

分享0 收藏0 回帖

關(guān)鍵詞：工具變量 性工具 調(diào)諧 regularization 測試 經(jīng)驗(yàn) sample