Alibaba

Things worth keeping.

A running collection in three shelves: the research papers I keep returning to (many from my master's thesis on sparse regression at BNP Paribas AM), general resources, and the interview-prep material that actually helped. Updated whenever something deserves a place.

01Research papers - 22
arXiv

Volatility is rough

Volatility is rough: the H=0.1 finding behind the rough-volatility deep dive on this site.

Gatheral, Jaisson, Rosenbaum · 2014
arXiv

Deep Hedging

Reframes hedging as an optimisation problem solvable with deep learning. Opens up an entire research direction.

Buehler, Gonon, Teichmann, Wood · 2019
arXiv / NeurIPS

Transfer Learning via L1 Regularization

The transfer-Lasso idea behind the fund risk-exposure pipeline I built at BNP Paribas AM.

Takada, Fujisawa · 2020
preprint

Adaptive Lasso, Transfer Lasso, and Beyond: An Asymptotic Perspective

Asymptotics unifying the adaptive and transfer Lasso estimators.

Takada, Fujisawa · 2024
JASA

The Adaptive Lasso and Its Oracle Properties

Weighted L1 penalties recover the oracle: the reference point for every Lasso variant that followed.

Zou · 2006
Ann. Statist.

Least Angle Regression

Efron, Hastie, Johnstone, Tibshirani · 2004
JMLR

On Model Selection Consistency of Lasso

The irrepresentable condition: when the Lasso can and cannot find the true support.

Zhao, Yu · 2006
JASA

Variable Selection via Nonconcave Penalized Likelihood and its Oracle Properties

SCAD: the nonconvex penalty that fixes the Lasso's bias on large coefficients.

Fan, Li · 2001
arXiv / Ann. Statist.

Near-ideal Model Selection by L1 Minimization

Candes, Plan · 2009
Ann. Statist.

On the Degrees of Freedom of the Lasso

Zou, Hastie, Tibshirani · 2007
CSDA

Relaxed Lasso

Meinshausen · 2007
Ann. Statist.

The Sparsity and Bias of the Lasso Selection in High-Dimensional Linear Regression

Zhang, Huang · 2008
Electron. J. Statist.

Thresholding-based Iterative Selection Procedures for Model Selection and Shrinkage

She · 2009
Machine Learning

Gene Selection for Cancer Classification using Support Vector Machines

SVM-RFE: recursive feature elimination, still a workhorse for feature selection.

Guyon, Weston, Barnhill, Vapnik · 2002
Statist. Surveys

Wavelet Methods in Statistics: Some Recent Developments and Their Applications

Antoniadis · 2007
Neural Computation

Comparison of Model Selection for Regression

Cherkassky, Ma · 2003
IEEE ISIT

Model Selection: Two Fundamental Measures of Coherence and Their Algorithmic Significance

Bajwa, Calderbank, Jafarpour · 2010
preprint

A Feature Selection Method for Multi-Dimension Time-Series Data

Kathirgamanathan, Cunningham · 2021
IEEE TSP

Dynamic Filtering of Time-Varying Sparse Signals via L1 Minimization

Where sparse recovery meets Kalman filtering: tracking a moving support.

Charles, Balavoine, Rozell · 2016
preprint

A Step by Step Mathematical Derivation and Tutorial on Kalman Filters

The cleanest self-contained derivation of the Kalman filter I know.

Masnadi-Shirazi, Masnadi-Shirazi, Dastgheib · 2019
SSRN

Return-Based Style Analysis with Time-Varying Exposures

Kalman-filtered style exposures: the dynamic version of Sharpe's style analysis.

Swinkels, Van der Sluis · 2002
MSCI Barra

Extreme Risk Analysis

Goldberg, Miller, Weinstein · 2009
02Resources - books, sites, education
03Interview preparation