具有錯誤發現率和型一誤差控制的可解釋之預測樹模型

A tree-based interpretable predictive method with FDR and type-one error control

洪晟恩(Cheng-En Hong)

指導教授 ： 歐陽彥正

國立臺灣大學/共同教育中心/統計碩士學位學程/碩士(2017年)

https://doi.org/10.6342/NTU201702789

摘要

在實際的問題中儘管擁有許多的變數，我們並不曉得哪些變數是
真實的變數，哪些是虛假的雜訊。通過發現重要變數，研究人員可以
進一步利用選擇的重要變數進行更有針對性的後續實驗以利探討背
後的科學現象。一個自然的要求是，我們希望盡可能發現更多的相
關變量，同時盡可能犯更少的錯誤。我們提出一個改良的RuleFit 模
型，其中包含利用knockoff procedure 達到控制錯誤發現率, 以及通過
Neyman-Pearson 方法控制型一誤差。

關鍵字

模型選擇 ； 錯誤發現率

並列摘要

Despite the abundance of the available variables, ground truth is privy
to knowledge about the problem seldom revealed in practice. By discovering
important features, researchers can further conduct a more targeted follow-up
experiment on the selected features tailored for understanding the scientific
phenomenon. A natural requirement is that we wish to discover as many relevant
variables as possible and make as few mistakes as possible at the same
time. We propose a modified RuleFit with FDR control by knockoff procedure
and with alpha control by Neyman-Pearson method.

並列關鍵字

Knockoff ； FDR ； Lasso ； Neyman-Pearson method

參考文獻

[3] Barber, R. F., Candès, E. J., et al. (2015). Controlling the false discovery rate via knockoffs. The Annals of Statistics, 43(5):2055–2085.

[4] Beauchamp, N. (2017). Predicting and interpolating state-level polls using twitter textual data. American Journal of Political Science, 61(2):490–503.

[7] Bertsimas, D. and Dunn, J. (2017). Optimal classification trees. Machine Learning, 106(7):1–44.

[10] Breiman, L. and Shang, N. (1996). Born again trees. University of California, Berkeley, Berkeley, CA, Technical Report.

[11] Brzyski, D., Peterson, C. B., Sobczyk, P., Candes, E. J., Bogdan, M., and Sabatti, C. (2017). Controlling the rate of gwas false discoveries. Genetics, 205(1):61–75.