arXiv subjects

Yuya Sasaki

Publications and source records attributed to Yuya Sasaki.

At least 55 records · Page 3Linked to original sources

arXiv2023

Extreme Changes in Changes

Policy analysts are often interested in treating the units with extreme outcomes, such as infants with extremely low birth weights. Existing changes-in-changes (CIC) estimators are tailored to middle quantiles and do not work well for such subpopulations. This paper proposes a new CIC estimator to accurately estimate treatment effects at extreme quantiles. With its asymptotic normality, we also propose a method of statistical inference, which is simple to implement. Based on simulation studies, we propose to use our extreme CIC estimator for extreme, such as below 5% and above 95%, quantiles, while the conventional CIC estimator should be used for intermediate quantiles. Applying the proposed method, we study the effects of income gains from the 1993 EITC reform on infant birth weights for those in the most critical conditions. This paper is accompanied by a Stata command.

econ.EM↗

arXiv2023

Tuning Parameter-Free Nonparametric Density Estimation from Tabulated Summary Data

Administrative data are often easier to access as tabulated summaries than in the original format due to confidentiality concerns. Motivated by this practical feature, we propose a novel nonparametric density estimation method from tabulated summary data based on maximum entropy and prove its strong uniform consistency. Unlike existing kernel-based estimators, our estimator is free from tuning parameters and admits a closed-form density that is convenient for post-estimation analysis. We apply the proposed method to the tabulated summary data of the U.S. tax returns to estimate the income distribution.

econ.EM↗

arXiv2023

Scardina: Scalable Join Cardinality Estimation by Multiple Density Estimators

In recent years, machine learning-based cardinality estimation methods are replacing traditional methods. This change is expected to contribute to one of the most important applications of cardinality estimation, the query optimizer, to speed up query processing. However, none of the existing methods do not precisely estimate cardinalities when relational schemas consist of many tables with strong correlations between tables/attributes. This paper describes that multiple density estimators can be combined to effectively target the cardinality estimation of data with large and complex schemas having strong correlations. We propose Scardina, a new join cardinality estimation method using multiple partitioned models based on the schema structure.

cs.DB↗

arXiv2023

NoSQL Schema Design for Time-Dependent Workloads

In this paper, we propose a schema optimization method for time-dependent workloads for NoSQL databases. In our proposed method, we migrate schema according to changing workloads, and the estimated cost of execution and migration are formulated and minimized as a single integer linear programming problem. Furthermore, we propose a method to reduce the number of optimization candidates by iterating over the time dimension abstraction and optimizing the workload while updating constraints.

cs.DB↗

arXiv2023

GenCAT: Generating Attributed Graphs with Controlled Relationships between Classes, Attributes, and Topology

Generating large synthetic attributed graphs with node labels is an important task to support various experimental studies for graph analysis methods. Existing graph generators fail to simultaneously simulate the relationships between labels, attributes, and topology which real-world graphs exhibit. Motivated by this limitation, we propose GenCAT, an attributed graph generator for controlling those relationships, which has the following advantages. (i) GenCAT generates graphs with user-specified node degrees and flexibly controls the relationship between nodes and labels by incorporating the connection proportion for each node to classes. (ii) Generated attribute values follow user-specified distributions, and users can flexibly control the correlation between the attributes and labels. (iii) Graph generation scales linearly to the number of edges. GenCAT is the first generator to support all three of these practical features. Through extensive experiments, we demonstrate that GenCAT can efficiently generate high-quality complex attributed graphs with user-controlled relationships between labels, attributes, and topology.

cs.SI↗

arXiv2023

On Using The Two-Way Cluster-Robust Standard Errors

Thousands of papers have reported two-way cluster-robust (TWCR) standard errors. However, the recent econometrics literature points out the potential non-gaussianity of two-way cluster sample means, and thus invalidity of the inference based on the TWCR standard errors. Fortunately, simulation studies nonetheless show that the gaussianity is rather common than exceptional. This paper provides theoretical support for this encouraging observation. Specifically, we derive a novel central limit theorem for two-way clustered triangular arrays that justifies the use of the TWCR under very mild and interpretable conditions. We, therefore, hope that this paper will provide a theoretical justification for the legitimacy of most, if not all, of the thousands of those empirical papers that have used the TWCR standard errors. We provide a guide in practice as to when a researcher can employ the TWCR standard errors.

econ.EM↗

arXiv2022

Inference in high-dimensional regression models without the exact or $L^p$ sparsity

This paper proposes a new method of inference in high-dimensional regression models and high-dimensional IV regression models. Estimation is based on a combined use of the orthogonal greedy algorithm, high-dimensional Akaike information criterion, and double/debiased machine learning. The method of inference for any low-dimensional subvector of high-dimensional parameters is based on a root-$N$ asymptotic normality, which is shown to hold without requiring the exact sparsity condition or the $L^p$ sparsity condition. Simulation studies demonstrate superior finite-sample performance of this proposed method over those based on the LASSO or the random forest, especially under less sparse models. We illustrate an application to production analysis with a panel of Chilean firms.

econ.EM↗

arXiv2022

Beyond Real-world Benchmark Datasets: An Empirical Study of Node Classification with GNNs

Graph Neural Networks (GNNs) have achieved great success on a node classification task. Despite the broad interest in developing and evaluating GNNs, they have been assessed with limited benchmark datasets. As a result, the existing evaluation of GNNs lacks fine-grained analysis from various characteristics of graphs. Motivated by this, we conduct extensive experiments with a synthetic graph generator that can generate graphs having controlled characteristics for fine-grained analysis. Our empirical studies clarify the strengths and weaknesses of GNNs from four major characteristics of real-world graphs with class labels of nodes, i.e., 1) class size distributions (balanced vs. imbalanced), 2) edge connection proportions between classes (homophilic vs. heterophilic), 3) attribute values (biased vs. random), and 4) graph sizes (small vs. large). In addition, to foster future research on GNNs, we publicly release our codebase that allows users to evaluate various GNNs with various graphs. We hope this work offers interesting insights for future research.

cs.LG↗

arXiv2022

Dyadic double/debiased machine learning for analyzing determinants of free trade agreements

This paper presents novel methods and theories for estimation and inference about parameters in econometric models using machine learning for nuisance parameters estimation when data are dyadic. We propose a dyadic cross fitting method to remove over-fitting biases under arbitrary dyadic dependence. Together with the use of Neyman orthogonal scores, this novel cross fitting method enables root-$n$ consistent estimation and inference robustly against dyadic dependence. We illustrate an application of our general framework to high-dimensional network link formation models. With this method applied to empirical data of international economic networks, we reexamine determinants of free trade agreements (FTA) viewed as links formed in the dyad composed of world economies. We document that standard methods may lead to misleading conclusions for numerous classic determinants of FTA formation due to biased point estimates or standard errors which are too small.

econ.EM↗

arXiv2022

Algorithmic subsampling under multiway clustering

This paper proposes a novel method of algorithmic subsampling (data sketching) for multiway cluster dependent data. We establish a new uniform weak law of large numbers and a new central limit theorem for the multiway algorithmic subsample means. Consequently, we discover an additional advantage of the algorithmic subsampling that it allows for robustness against potential degeneracy, and even non-Gaussian degeneracy, of the asymptotic distribution under multiway clustering. Simulation studies support this novel result, and demonstrate that inference with the algorithmic subsampling entails more accuracy than that without the algorithmic subsampling. Applying these basic asymptotic theories, we derive the consistency and the asymptotic normality for the multiway algorithmic subsampling generalized method of moments estimator and for the multiway algorithmic subsampling M-estimator. We illustrate an application to scanner data.

econ.EM↗

arXiv2022

Path association rule mining

Graph association rule mining is a data mining technique used for discovering regularities in graph data. In this study, we propose a novel concept, {\it path association rule mining}, to discover the correlations of path patterns that frequently appear in a given graph. Reachability path patterns (i.e., existence of paths from a vertex to another vertex) are applied in our concept to discover diverse regularities. We show that the problem is NP-hard, and we develop an efficient algorithm in which the anti-monotonic property is used on path patterns. Subsequently, we develop approximation and parallelization techniques to efficiently and scalably discover rules. We use real-life graphs to experimentally verify the effective

cs.DB↗

arXiv2022

On Uniform Confidence Intervals for the Tail Index and the Extreme Quantile

This paper presents two results concerning uniform confidence intervals for the tail index and the extreme quantile. First, we show that it is impossible to construct a length-optimal confidence interval satisfying the correct uniform coverage over a local non-parametric family of tail distributions. Second, in light of the impossibility result, we construct honest confidence intervals that are uniformly valid by incorporating the worst-case bias in the local non-parametric family. The proposed method is applied to simulated data and a real data set of National Vital Statistics from National Center for Health Statistics.

math.ST↗

arXiv2022

Fixed-k Tail Regression: New Evidence on Tax and Wealth Inequality from Forbes 400

We develop a novel fixed-k tail regression method that accommodates the unique feature in the Forbes 400 data that observations are truncated from below at the 400th largest order statistic. Applying this method, we find that higher maximum marginal income tax rates induce higher wealth Pareto exponents. Setting the maximum tax rate to 30-40% (as in U.S. currently) leads to a Pareto exponent of 1.5-1.8, while counterfactually setting it to 80% (as suggested by Piketty, 2014) would lead to a Pareto exponent of 2.6. We present a simple economic model that explains these findings and discuss the welfare implications of taxation.

econ.GN↗

arXiv2022

Estimation of Average Derivatives of Latent Regressors: With an Application to Inference on Buffer-Stock Saving

This paper proposes a density-weighted average derivative estimator based on two noisy measures of a latent regressor. Both measures have classical errors with possibly asymmetric distributions. We show that the proposed estimator achieves the root-n rate of convergence, and derive its asymptotic normal distribution for statistical inference. Simulation studies demonstrate excellent small-sample performance supporting the root-n asymptotic normality. Based on the proposed estimator, we construct a formal test on the sub-unity of the marginal propensity to consume out of permanent income (MPCP) under a nonparametric consumption model and a permanent-transitory model of income dynamics with nonparametric distribution. Applying the test to four recent waves of U.S. Panel Study of Income Dynamics (PSID), we reject the null hypothesis of the unit MPCP in favor of a sub-unit MPCP, supporting the buffer-stock model of saving.

econ.EM↗

arXiv2022

GNN Transformation Framework for Improving Efficiency and Scalability

We propose a framework that automatically transforms non-scalable GNNs into precomputation-based GNNs which are efficient and scalable for large-scale graphs. The advantages of our framework are two-fold; 1) it transforms various non-scalable GNNs to scale well to large-scale graphs by separating local feature aggregation from weight learning in their graph convolution, 2) it efficiently executes precomputation on GPU for large-scale graphs by decomposing their edges into small disjoint and balanced sets. Through extensive experiments with large-scale graphs, we demonstrate that the transformed GNNs run faster in training time than existing GNNs while achieving competitive accuracy to the state-of-the-art GNNs. Consequently, our transformation framework provides simple and efficient baselines for future research on scalable GNNs.

cs.LG↗

arXiv2022

Scaling Private Deep Learning with Low-Rank and Sparse Gradients

Applying Differentially Private Stochastic Gradient Descent (DPSGD) to training modern, large-scale neural networks such as transformer-based models is a challenging task, as the magnitude of noise added to the gradients at each iteration scales with model dimension, hindering the learning capability significantly. We propose a unified framework, $\textsf{LSG}$, that fully exploits the low-rank and sparse structure of neural networks to reduce the dimension of gradient updates, and hence alleviate the negative impacts of DPSGD. The gradient updates are first approximated with a pair of low-rank matrices. Then, a novel strategy is utilized to sparsify the gradients, resulting in low-dimensional, less noisy updates that are yet capable of retaining the performance of neural networks. Empirical evaluation on natural language processing and computer vision tasks shows that our method outperforms other state-of-the-art baselines.

cs.LG↗

arXiv2022

An Empirical Study of Personalized Federated Learning

Federated learning is a distributed machine learning approach in which a single server and multiple clients collaboratively build machine learning models without sharing datasets on clients. A challenging issue of federated learning is data heterogeneity (i.e., data distributions may differ across clients). To cope with this issue, numerous federated learning methods aim at personalized federated learning and build optimized models for clients. Whereas existing studies empirically evaluated their own methods, the experimental settings (e.g., comparison methods, datasets, and client setting) in these studies differ from each other, and it is unclear which personalized federate learning method achieves the best performance and how much progress can be made by using these methods instead of standard (i.e., non-personalized) federated learning. In this paper, we benchmark the performance of existing personalized federated learning through comprehensive experiments to evaluate the characteristics of each method. Our experimental study shows that (1) there are no champion methods, (2) large data heterogeneity often leads to high accurate predictions, and (3) standard federated learning methods (e.g. FedAvg) with fine-tuning often outperform personalized federated learning methods. We open our benchmark tool FedBench for researchers to conduct experimental studies with various experimental settings.

cs.LG↗

arXiv2022

Predicting Parking Lot Availability by Graph-to-Sequence Model: A Case Study with SmartSantander

Nowadays, so as to improve services and urban areas livability, multiple smart city initiatives are being carried out throughout the world. SmartSantander is a smart city project in Santander, Spain, which has relied on wireless sensor network technologies to deploy heterogeneous sensors within the city to measure multiple parameters, including outdoor parking information. In this paper, we study the prediction of parking lot availability using historical data from more than 300 outdoor parking sensors with SmartSantander. We design a graph-to-sequence model to capture the periodical fluctuation and geographical proximity of parking lots. For developing and evaluating our model, we use a 3-year dataset of parking lot availability in the city of Santander. Our model achieves a high accuracy compared with existing sequence-to-sequence models, which is accurate enough to provide a parking information service in the city. We apply our model to a smartphone application to be widely used by citizens and tourists.

cs.LG↗