High-dimensional matrix-valued time series are of significant interest in economics and finance, with prominent examples including cross region macroeconomic panels and firms' financial data panels. We introduce a class of Bayesian matrix dynamic factor models that utilize matrix structures to identify more interpretable factor patterns and factor impacts. Our model accommodates time-varying volatility, adjusts for outliers, and allows cross-sectional correlations in the idiosyncratic components. To determine the dimension of the factor matrix, we employ an importance-sampling estimator based on the cross-entropy method to estimate marginal likelihoods. Through a series of Monte Carlo experiments, we show the properties of the factor estimators and the performance of the marginal likelihood estimator in correctly identifying the true dimensions of the factor matrices. Applying our model to a macroeconomic dataset and a financial dataset, we demonstrate its ability in unveiling interesting features within matrix-valued time series.
{"title":"Bayesian Dynamic Factor Models for High-dimensional Matrix-valued Time Series","authors":"Wei Zhang","doi":"arxiv-2409.08354","DOIUrl":"https://doi.org/arxiv-2409.08354","url":null,"abstract":"High-dimensional matrix-valued time series are of significant interest in\u0000economics and finance, with prominent examples including cross region\u0000macroeconomic panels and firms' financial data panels. We introduce a class of\u0000Bayesian matrix dynamic factor models that utilize matrix structures to\u0000identify more interpretable factor patterns and factor impacts. Our model\u0000accommodates time-varying volatility, adjusts for outliers, and allows\u0000cross-sectional correlations in the idiosyncratic components. To determine the\u0000dimension of the factor matrix, we employ an importance-sampling estimator\u0000based on the cross-entropy method to estimate marginal likelihoods. Through a\u0000series of Monte Carlo experiments, we show the properties of the factor\u0000estimators and the performance of the marginal likelihood estimator in\u0000correctly identifying the true dimensions of the factor matrices. Applying our\u0000model to a macroeconomic dataset and a financial dataset, we demonstrate its\u0000ability in unveiling interesting features within matrix-valued time series.","PeriodicalId":501293,"journal":{"name":"arXiv - ECON - Econometrics","volume":"210 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142261055","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We propose sieve wild bootstrap analogues to the adaptive Lasso solution path unit root tests of Arnold and Reinschl"ussel (2024) arXiv:2404.06205 to improve finite sample properties and extend their applicability to a generalised framework, allowing for non-stationary volatility. Numerical evidence shows the bootstrap to improve the tests' precision for error processes that promote spurious rejections of the unit root null, depending on the detrending procedure. The bootstrap mitigates finite-sample size distortions and restores asymptotically valid inference when the data features time-varying unconditional variance. We apply the bootstrap tests to real residential property prices of the top six Eurozone economies and find evidence of stationarity to be period-specific, supporting the conjecture that exuberance in the housing market characterises the development of Euro-era residential property prices in the recent past.
我们对 Arnold 和 Reinschl"ussel (2024) 的自适应拉索解路径单位根检验提出了筛子式自举类似方法，以改进有限样本属性，并将其适用性扩展到广义框架，允许非平稳波动。数值证据表明，自举法提高了检验的精确度，因为误差过程会导致对单位根零值的虚假拒绝，这取决于去趋势过程。当数据的无条件方差随时间变化时，自举法可减轻有限样本的偏差并恢复渐近有效的推断。我们对欧元区前六大经济体的实际住宅物业价格进行了自举检验，发现静止性的证据是特定时期的，这支持了住房市场的繁荣是近期欧元区住宅物业价格发展特点的猜想。
{"title":"Bootstrap Adaptive Lasso Solution Path Unit Root Tests","authors":"Martin C. Arnold, Thilo Reinschlüssel","doi":"arxiv-2409.07859","DOIUrl":"https://doi.org/arxiv-2409.07859","url":null,"abstract":"We propose sieve wild bootstrap analogues to the adaptive Lasso solution path\u0000unit root tests of Arnold and Reinschl\"ussel (2024) arXiv:2404.06205 to\u0000improve finite sample properties and extend their applicability to a\u0000generalised framework, allowing for non-stationary volatility. Numerical\u0000evidence shows the bootstrap to improve the tests' precision for error\u0000processes that promote spurious rejections of the unit root null, depending on\u0000the detrending procedure. The bootstrap mitigates finite-sample size\u0000distortions and restores asymptotically valid inference when the data features\u0000time-varying unconditional variance. We apply the bootstrap tests to real\u0000residential property prices of the top six Eurozone economies and find evidence\u0000of stationarity to be period-specific, supporting the conjecture that\u0000exuberance in the housing market characterises the development of Euro-era\u0000residential property prices in the recent past.","PeriodicalId":501293,"journal":{"name":"arXiv - ECON - Econometrics","volume":"1 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142184088","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Mogens Fosgerau, Nikolaj Nielsen, Mads Paulsen, Thomas Kjær Rasmussen, Rui Yao
This paper considers substitution patterns in the perturbed utility route choice model. We provide a general result that determines the marginal change in link flows following a marginal change in link costs across the network. We give a general condition on the network structure under which all paths are necessarily substitutes and an example in which some paths are complements. The presence of complementarity contradicts a result in a previous paper in this journal; we point out and correct the error.
{"title":"Substitution in the perturbed utility route choice model","authors":"Mogens Fosgerau, Nikolaj Nielsen, Mads Paulsen, Thomas Kjær Rasmussen, Rui Yao","doi":"arxiv-2409.08347","DOIUrl":"https://doi.org/arxiv-2409.08347","url":null,"abstract":"This paper considers substitution patterns in the perturbed utility route\u0000choice model. We provide a general result that determines the marginal change\u0000in link flows following a marginal change in link costs across the network. We\u0000give a general condition on the network structure under which all paths are\u0000necessarily substitutes and an example in which some paths are complements. The\u0000presence of complementarity contradicts a result in a previous paper in this\u0000journal; we point out and correct the error.","PeriodicalId":501293,"journal":{"name":"arXiv - ECON - Econometrics","volume":"4 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142261056","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
We propose a test to detect a forecast accuracy breakdown in a long memory time series and provide theoretical and simulation evidence on the memory transfer from the time series to the forecast residuals. The proposed method uses a double sup-Wald test against the alternative of a structural break in the mean of an out-of-sample loss series. To address the problem of estimating the long-run variance under long memory, a robust estimator is applied. The corresponding breakpoint results from a long memory robust CUSUM test. The finite sample size and power properties of the test are derived in a Monte Carlo simulation. A monotonic power function is obtained for the fixed forecasting scheme. In our practical application, we find that the global energy crisis that began in 2021 led to a forecast break in European electricity prices, while the results for the U.S. are mixed.
{"title":"Testing for a Forecast Accuracy Breakdown under Long Memory","authors":"Jannik Kreye, Philipp Sibbertsen","doi":"arxiv-2409.07087","DOIUrl":"https://doi.org/arxiv-2409.07087","url":null,"abstract":"We propose a test to detect a forecast accuracy breakdown in a long memory\u0000time series and provide theoretical and simulation evidence on the memory\u0000transfer from the time series to the forecast residuals. The proposed method\u0000uses a double sup-Wald test against the alternative of a structural break in\u0000the mean of an out-of-sample loss series. To address the problem of estimating\u0000the long-run variance under long memory, a robust estimator is applied. The\u0000corresponding breakpoint results from a long memory robust CUSUM test. The\u0000finite sample size and power properties of the test are derived in a Monte\u0000Carlo simulation. A monotonic power function is obtained for the fixed\u0000forecasting scheme. In our practical application, we find that the global\u0000energy crisis that began in 2021 led to a forecast break in European\u0000electricity prices, while the results for the U.S. are mixed.","PeriodicalId":501293,"journal":{"name":"arXiv - ECON - Econometrics","volume":"11 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142184089","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper proposes methods of estimation and uniform inference for a general class of causal functions, such as the conditional average treatment effects and the continuous treatment effects, under multiway clustering. The causal function is identified as a conditional expectation of an adjusted (Neyman-orthogonal) signal that depends on high-dimensional nuisance parameters. We propose a two-step procedure where the first step uses machine learning to estimate the high-dimensional nuisance parameters. The second step projects the estimated Neyman-orthogonal signal onto a dictionary of basis functions whose dimension grows with the sample size. For this two-step procedure, we propose both the full-sample and the multiway cross-fitting estimation approaches. A functional limit theory is derived for these estimators. To construct the uniform confidence bands, we develop a novel resampling procedure, called the multiway cluster-robust sieve score bootstrap, that extends the sieve score bootstrap (Chen and Christensen, 2018) to the novel setting with multiway clustering. Extensive numerical simulations showcase that our methods achieve desirable finite-sample behaviors. We apply the proposed methods to analyze the causal relationship between mistrust levels in Africa and the historical slave trade. Our analysis rejects the null hypothesis of uniformly zero effects and reveals heterogeneous treatment effects, with significant impacts at higher levels of trade volumes.
{"title":"Estimation and Inference for Causal Functions with Multiway Clustered Data","authors":"Nan Liu, Yanbo Liu, Yuya Sasaki","doi":"arxiv-2409.06654","DOIUrl":"https://doi.org/arxiv-2409.06654","url":null,"abstract":"This paper proposes methods of estimation and uniform inference for a general\u0000class of causal functions, such as the conditional average treatment effects\u0000and the continuous treatment effects, under multiway clustering. The causal\u0000function is identified as a conditional expectation of an adjusted\u0000(Neyman-orthogonal) signal that depends on high-dimensional nuisance\u0000parameters. We propose a two-step procedure where the first step uses machine\u0000learning to estimate the high-dimensional nuisance parameters. The second step\u0000projects the estimated Neyman-orthogonal signal onto a dictionary of basis\u0000functions whose dimension grows with the sample size. For this two-step\u0000procedure, we propose both the full-sample and the multiway cross-fitting\u0000estimation approaches. A functional limit theory is derived for these\u0000estimators. To construct the uniform confidence bands, we develop a novel\u0000resampling procedure, called the multiway cluster-robust sieve score bootstrap,\u0000that extends the sieve score bootstrap (Chen and Christensen, 2018) to the\u0000novel setting with multiway clustering. Extensive numerical simulations\u0000showcase that our methods achieve desirable finite-sample behaviors. We apply\u0000the proposed methods to analyze the causal relationship between mistrust levels\u0000in Africa and the historical slave trade. Our analysis rejects the null\u0000hypothesis of uniformly zero effects and reveals heterogeneous treatment\u0000effects, with significant impacts at higher levels of trade volumes.","PeriodicalId":501293,"journal":{"name":"arXiv - ECON - Econometrics","volume":"24 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142184090","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Shen Li, Yuyang Zhang, Zhaolin Ren, Claire Liang, Na Li, Julie A. Shah
Binary human choice feedback is widely used in interactive preference learning for its simplicity, but it provides limited information about preference strength. To overcome this limitation, we leverage human response times, which inversely correlate with preference strength, as complementary information. Our work integrates the EZ-diffusion model, which jointly models human choices and response times, into preference-based linear bandits. We introduce a computationally efficient utility estimator that reformulates the utility estimation problem using both choices and response times as a linear regression problem. Theoretical and empirical comparisons with traditional choice-only estimators reveal that for queries with strong preferences ("easy" queries), choices alone provide limited information, while response times offer valuable complementary information about preference strength. As a result, incorporating response times makes easy queries more useful. We demonstrate this advantage in the fixed-budget best-arm identification problem, with simulations based on three real-world datasets, consistently showing accelerated learning when response times are incorporated.
二进制人类选择反馈因其简单性被广泛应用于交互式偏好学习中，但它提供的偏好强度信息有限。为了克服这一局限，我们利用与偏好强度成反比的人类反应时间作为补充信息。我们的工作将 EZ 扩散模型与基于偏好的线性匪帮模型相结合，EZ 扩散模型可以对人类的选择和响应时间进行联合建模。我们引入了一种计算效率高的效用估计器，它将使用选择和响应时间的效用估计问题重新表述为线性回归问题。通过与传统的仅有选择的估计器进行理论和实证比较，我们发现对于具有强烈偏好的查询（"简单 "查询），仅有选择提供的信息是有限的，而响应时间则提供了关于偏好强度的宝贵补充信息。因此，加入响应时间会使简单查询更有用。我们在固定预算最佳臂识别问题中证明了这一优势，并基于三个真实世界数据集进行了模拟。
{"title":"Enhancing Preference-based Linear Bandits via Human Response Time","authors":"Shen Li, Yuyang Zhang, Zhaolin Ren, Claire Liang, Na Li, Julie A. Shah","doi":"arxiv-2409.05798","DOIUrl":"https://doi.org/arxiv-2409.05798","url":null,"abstract":"Binary human choice feedback is widely used in interactive preference\u0000learning for its simplicity, but it provides limited information about\u0000preference strength. To overcome this limitation, we leverage human response\u0000times, which inversely correlate with preference strength, as complementary\u0000information. Our work integrates the EZ-diffusion model, which jointly models\u0000human choices and response times, into preference-based linear bandits. We\u0000introduce a computationally efficient utility estimator that reformulates the\u0000utility estimation problem using both choices and response times as a linear\u0000regression problem. Theoretical and empirical comparisons with traditional\u0000choice-only estimators reveal that for queries with strong preferences (\"easy\"\u0000queries), choices alone provide limited information, while response times offer\u0000valuable complementary information about preference strength. As a result,\u0000incorporating response times makes easy queries more useful. We demonstrate\u0000this advantage in the fixed-budget best-arm identification problem, with\u0000simulations based on three real-world datasets, consistently showing\u0000accelerated learning when response times are incorporated.","PeriodicalId":501293,"journal":{"name":"arXiv - ECON - Econometrics","volume":"34 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142184096","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper presents uniform estimation and inference theory for a large class of nonparametric partitioning-based M-estimators. The main theoretical results include: (i) uniform consistency for convex and non-convex objective functions; (ii) optimal uniform Bahadur representations; (iii) optimal uniform (and mean square) convergence rates; (iv) valid strong approximations and feasible uniform inference methods; and (v) extensions to functional transformations of underlying estimators. Uniformity is established over both the evaluation point of the nonparametric functional parameter and a Euclidean parameter indexing the class of loss functions. The results also account explicitly for the smoothness degree of the loss function (if any), and allow for a possibly non-identity (inverse) link function. We illustrate the main theoretical and methodological results with four substantive applications: quantile regression, distribution regression, $L_p$ regression, and Logistic regression; many other possibly non-smooth, nonlinear, generalized, robust M-estimation settings are covered by our theoretical results. We provide detailed comparisons with the existing literature and demonstrate substantive improvements: we achieve the best (in some cases optimal) known results under improved (in some cases minimal) requirements in terms of regularity conditions and side rate restrictions. The supplemental appendix reports other technical results that may be of independent interest.
{"title":"Uniform Estimation and Inference for Nonparametric Partitioning-Based M-Estimators","authors":"Matias D. Cattaneo, Yingjie Feng, Boris Shigida","doi":"arxiv-2409.05715","DOIUrl":"https://doi.org/arxiv-2409.05715","url":null,"abstract":"This paper presents uniform estimation and inference theory for a large class\u0000of nonparametric partitioning-based M-estimators. The main theoretical results\u0000include: (i) uniform consistency for convex and non-convex objective functions;\u0000(ii) optimal uniform Bahadur representations; (iii) optimal uniform (and mean\u0000square) convergence rates; (iv) valid strong approximations and feasible\u0000uniform inference methods; and (v) extensions to functional transformations of\u0000underlying estimators. Uniformity is established over both the evaluation point\u0000of the nonparametric functional parameter and a Euclidean parameter indexing\u0000the class of loss functions. The results also account explicitly for the\u0000smoothness degree of the loss function (if any), and allow for a possibly\u0000non-identity (inverse) link function. We illustrate the main theoretical and\u0000methodological results with four substantive applications: quantile regression,\u0000distribution regression, $L_p$ regression, and Logistic regression; many other\u0000possibly non-smooth, nonlinear, generalized, robust M-estimation settings are\u0000covered by our theoretical results. We provide detailed comparisons with the\u0000existing literature and demonstrate substantive improvements: we achieve the\u0000best (in some cases optimal) known results under improved (in some cases\u0000minimal) requirements in terms of regularity conditions and side rate\u0000restrictions. The supplemental appendix reports other technical results that\u0000may be of independent interest.","PeriodicalId":501293,"journal":{"name":"arXiv - ECON - Econometrics","volume":"26 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142184097","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Partial least squares (PLS) is a simple factorisation method that works well with high dimensional problems in which the number of observations is limited given the number of independent variables. In this article, we show that PLS can perform better than ordinary least squares (OLS), least absolute shrinkage and selection operator (LASSO) and ridge regression in forecasting quarterly gross domestic product (GDP) growth, covering the period from 2000 to 2023. In fact, through dimension reduction, PLS proved to be effective in lowering the out-of-sample forecasting error, specially since 2020. For the period 2000-2019, the four methods produce similar results, suggesting that PLS is a valid regularisation technique like LASSO or ridge.
{"title":"The Surprising Robustness of Partial Least Squares","authors":"João B. Assunção, Pedro Afonso Fernandes","doi":"arxiv-2409.05713","DOIUrl":"https://doi.org/arxiv-2409.05713","url":null,"abstract":"Partial least squares (PLS) is a simple factorisation method that works well\u0000with high dimensional problems in which the number of observations is limited\u0000given the number of independent variables. In this article, we show that PLS\u0000can perform better than ordinary least squares (OLS), least absolute shrinkage\u0000and selection operator (LASSO) and ridge regression in forecasting quarterly\u0000gross domestic product (GDP) growth, covering the period from 2000 to 2023. In\u0000fact, through dimension reduction, PLS proved to be effective in lowering the\u0000out-of-sample forecasting error, specially since 2020. For the period\u00002000-2019, the four methods produce similar results, suggesting that PLS is a\u0000valid regularisation technique like LASSO or ridge.","PeriodicalId":501293,"journal":{"name":"arXiv - ECON - Econometrics","volume":"22 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142184091","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this study, we leverage powerful non-linear machine learning methods to identify the characteristics of trades that contain valuable information. First, we demonstrate the effectiveness of our optimized neural network predictor in accurately predicting future market movements. Then, we utilize the information from this successful neural network predictor to pinpoint the individual trades within each data point (trading window) that had the most impact on the optimized neural network's prediction of future price movements. This approach helps us uncover important insights about the heterogeneity in information content provided by trades of different sizes, venues, trading contexts, and over time.
{"title":"Bellwether Trades: Characteristics of Trades influential in Predicting Future Price Movements in Markets","authors":"Tejas Ramdas, Martin T. Wells","doi":"arxiv-2409.05192","DOIUrl":"https://doi.org/arxiv-2409.05192","url":null,"abstract":"In this study, we leverage powerful non-linear machine learning methods to\u0000identify the characteristics of trades that contain valuable information.\u0000First, we demonstrate the effectiveness of our optimized neural network\u0000predictor in accurately predicting future market movements. Then, we utilize\u0000the information from this successful neural network predictor to pinpoint the\u0000individual trades within each data point (trading window) that had the most\u0000impact on the optimized neural network's prediction of future price movements.\u0000This approach helps us uncover important insights about the heterogeneity in\u0000information content provided by trades of different sizes, venues, trading\u0000contexts, and over time.","PeriodicalId":501293,"journal":{"name":"arXiv - ECON - Econometrics","volume":"17 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142184098","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Confounding events with correlated timing violate the parallel trends assumption in Difference-in-Differences (DiD) designs. I show that the standard staggered DiD estimator is biased in the presence of confounding events. Identification can be achieved with units not yet treated by either event as controls and a double DiD design using variation in treatment timing. I apply this method to examine the effect of states' staggered minimum wage raise on teen employment from 2010 to 2020. The Medicaid expansion under the ACA confounded the raises, leading to a spurious negative estimate.
{"title":"Difference-in-Differences with Multiple Events","authors":"Lin-Tung Tsai","doi":"arxiv-2409.05184","DOIUrl":"https://doi.org/arxiv-2409.05184","url":null,"abstract":"Confounding events with correlated timing violate the parallel trends\u0000assumption in Difference-in-Differences (DiD) designs. I show that the standard\u0000staggered DiD estimator is biased in the presence of confounding events.\u0000Identification can be achieved with units not yet treated by either event as\u0000controls and a double DiD design using variation in treatment timing. I apply\u0000this method to examine the effect of states' staggered minimum wage raise on\u0000teen employment from 2010 to 2020. The Medicaid expansion under the ACA\u0000confounded the raises, leading to a spurious negative estimate.","PeriodicalId":501293,"journal":{"name":"arXiv - ECON - Econometrics","volume":"52 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142184092","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：