Policy Optimization Using Semiparametric Models for Dynamic Pricing

DOI10.1080/01621459.2022.2128359arXiv2109.06368OpenAlexW3199034510MaRDI QIDQ6154014FDOQ6154014

Authors: Jianqing Fan, Yongyi Guo 

Publication date: 19 March 2024

Published in: Journal of the American Statistical Association (Search for Journal in Brave)

Abstract: In this paper, we study the contextual dynamic pricing problem where the market value of a product is linear in its observed features plus some market noise. Products are sold one at a time, and only a binary response indicating success or failure of a sale is observed. Our model setting is similar to Javanmard and Nazerzadeh [2019] except that we expand the demand curve to a semiparametric model and need to learn dynamically both parametric and nonparametric components. We propose a dynamic statistical learning and decision-making policy that combines semiparametric estimation from a generalized linear model with an unknown link and online decision-making to minimize regret (maximize revenue). Under mild conditions, we show that for a market noise c.d.f. $F(cdot)$ with $m$-th order derivative ($mgeq2$), our policy achieves a regret upper bound of $ilde{O}_d(T^{frac2m+14m-1})$, where $T$ is time horizon and $ilde{O}_d$ is the order that hides logarithmic terms and the dimensionality of feature $d$. The upper bound is further reduced to $ilde{O}_d(sqrtT)$ if $F$ is super smooth whose Fourier transform decays exponentially. In terms of dependence on the horizon $T$, these upper bounds are close to $Omega(sqrtT)$, the lower bound where $F$ belongs to a parametric class. We further generalize these results to the case with dynamically dependent product features under the strong mixing condition.

Full work available at URL: https://arxiv.org/abs/2109.06368

zbMATH Keywords

nonparametric statisticspolicy optimizationcontextual dynamic pricinggeneralized linear model with unknown link

Mathematics Subject Classification ID

Statistics (62-XX)

Cites Work

• Semiparametric Regression
• Optimal smoothing in single-index models
• Title not available (Why is that?)
• Generalized linear models with unknown link functions
• Generalized Partially Linear Single-Index Models
• Title not available (Why is that?)
• Optimal rates of convergence for nonparametric estimators
• Semiparametric least squares (SLS) and weighted SLS estimation of single-index models
• Nonlinear time series. Nonparametric and parametric methods
• Direct Semiparametric Estimation of Single-Index Models with Discrete Covariates
• Semiparametric Estimation of Index Coefficients
• New Estimation and Model Selection Procedures for Semiparametric Modeling in Longitudinal Data Analysis
• Optimal global rates of convergence for nonparametric regression
• On the optimal rates of convergence for nonparametric deconvolution problems
• MULTIVARIATE LOCAL POLYNOMIAL REGRESSION FOR TIME SERIES:UNIFORM STRONG CONSISTENCY AND RATES
• A distribution-free theory of nonparametric regression
• Adapting for the missing link
• Weak and strong uniform consistency of kernel regression estimates
• On Single-Index Coefficient Regression Models
• Local Polynomial Kernel Regression for Generalized Linear Models and Quasi-Likelihood Functions
• An Efficient Semiparametric Estimator for Binary Response Models
• Weak and strong uniform consistency of the kernel estimate of a density and its derivatives
• Nonparametric regression with infinite order flat-top kernels
• Efficient estimation in conditional single-index regression
• Dynamic Pricing with an Unknown Demand Model: Asymptotically Optimal Semi-Myopic Policies
• Dynamic pricing without knowing the demand function: risk bounds and near-optimal algorithms
• Dynamic pricing under a general parametric choice model
• A Nonparametric Approach to Multiproduct Pricing
• CDF and survival function estimation with infinite-order kernels
• Close the gaps: a learning-while-doing algorithm for single-product revenue management problems
• Title not available (Why is that?)
• Title not available (Why is that?)
• Recent research developments of strategic consumer behavior in operations management
• Generalized Linear Models
• Infinite order cross-validated local polynomial regression
• Nonparametric Pricing Analytics with Customer Covariates
• Nonparametric Self-Adjusting Control for Joint Learning and Optimization of Multiproduct Pricing with Finite Resource Capacity
• Dynamic Incentive-Aware Learning: Robust Pricing in Contextual Auctions
• Title not available (Why is that?)

Cited In (1)

• COMBSS: best subset selection via continuous optimization

This page was built for publication: Policy Optimization Using Semiparametric Models for Dynamic Pricing

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6154014)