{"title":"Interpretable Machine Learning Using Partial Linear Models*","authors":"Emmanuel Flachaire, Sullivan Hué, Sébastien Laurent, Gilles Hacheme","doi":"10.1111/obes.12592","DOIUrl":"10.1111/obes.12592","url":null,"abstract":"<p>Despite their high predictive performance, random forest and gradient boosting are often considered as black boxes which has raised concerns from practitioners and regulators. As an alternative, we suggest using partial linear models that are inherently interpretable. Specifically, we propose to combine parametric and non-parametric functions to accurately capture linearities and non-linearities prevailing between dependent and explanatory variables, and a variable selection procedure to control for overfitting issues. Estimation relies on a two-step procedure building upon the double residual method. We illustrate the predictive performance and interpretability of our approach on a regression problem.</p>","PeriodicalId":54654,"journal":{"name":"Oxford Bulletin of Economics and Statistics","volume":"86 3","pages":"519-540"},"PeriodicalIF":2.5,"publicationDate":"2023-12-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/obes.12592","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139070216","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Impact of Graduating with Honours on Entry Wages of Economics Majors*","authors":"Salim Atay, Gunes A. Asik, Semih Tumen","doi":"10.1111/obes.12593","DOIUrl":"10.1111/obes.12593","url":null,"abstract":"<p>Employers use various proxies to predict the future labour productivity levels of the job applicants. Success in school, especially in high-level coursework, is among the most widely used proxies to screen entry-level candidates. We estimate the causal effect of graduating with honours (i.e. with a grade point average of 3.00 and above out of 4.00) on the starting wages of economics majors in Türkiye. Using comprehensive micro data on all economics majors between 2014 and 2018, matched with administrative records about their first jobs, we implement a regression discontinuity analysis to investigate whether there is any statistically significant jump in the starting wages at the honours-degree cutoff. We find that graduating with honours increases the wages of males, while there is no impact on females. We further document that the impact on males is almost entirely driven by the graduates of non-elite universities. In particular, graduating with an honours degree increases the entry wages of males from non-elite universities by about 4%, on average. We provide an explanation for these patterns using the theory of statistical discrimination. We discuss the potential reasons behind the heterogeneous signal value of graduating with honours between males vs. females and elite versus non-elite university graduates.</p>","PeriodicalId":54654,"journal":{"name":"Oxford Bulletin of Economics and Statistics","volume":"86 3","pages":"606-640"},"PeriodicalIF":2.5,"publicationDate":"2023-12-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/obes.12593","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139051425","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Value-at-Risk under Measurement Error","authors":"Mohamed Doukali, Xiaojun Song, Abderrahim Taamouti","doi":"10.1111/obes.12589","DOIUrl":"10.1111/obes.12589","url":null,"abstract":"<p>We propose a method for estimating Value-at-Risk that corrects for the effect of measurement errors in stock prices. We show that the presence of measurement errors might pose serious problems for estimating risk measures. In particular, when stock prices are contaminated, existing estimators of Value-at-Risk are inconsistent and might lead to an underestimation of risk, which can result in extreme leverage ratios within the held portfolios. Using a Fourier transform and a deconvolution kernel estimator of the probability distribution function of actual latent prices, we derive a robust estimator of Value-at-Risk in the presence of measurement errors. Monte Carlo simulations and real data analysis illustrate satisfactory performance of the proposed method.</p>","PeriodicalId":54654,"journal":{"name":"Oxford Bulletin of Economics and Statistics","volume":"86 3","pages":"690-713"},"PeriodicalIF":2.5,"publicationDate":"2023-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/obes.12589","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138576630","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Peer Migration in China","authors":"Yuyu Chen, Ginger Zhe Jin, Yang Yue","doi":"10.1111/obes.12588","DOIUrl":"10.1111/obes.12588","url":null,"abstract":"<p>With over 290 million rural labourers transitioning to urban areas in 2019, China is experiencing an unparalleled scale of internal migration, the largest in human history. Employing instrumental variables (IVs) gleaned from the 2006 China Agricultural Census (CAC), we find that a 10 percentage point increase in the migration rate among co-villagers amplifies an individual's probability of migrating by 7.13 percentage points. Influencing factors such as information dissemination at the origin and cost efficiencies at the destination likely contribute to the observed clustering of migration by age, destination and industrial sector. Intriguingly, migration seems to exert a negligible influence on the agricultural productivity of those who remain, which may be due to substantial labour redundancy at the point of origin and potentially higher productivity among migrants.</p>","PeriodicalId":54654,"journal":{"name":"Oxford Bulletin of Economics and Statistics","volume":"86 2","pages":"257-313"},"PeriodicalIF":2.5,"publicationDate":"2023-12-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138576647","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Solving the Forecast Combination Puzzle Using Double Shrinkages*","authors":"Li Liu, Xianfeng Hao, Yudong Wang","doi":"10.1111/obes.12590","DOIUrl":"10.1111/obes.12590","url":null,"abstract":"<p>This study develops a new approach that shrinks the forecast combination weights towards equal weights by using weighted least squares and towards zero weight by using regularization constraints. We reveal the significant predictability of excess returns to the S&P500 index that can be achieved by using this double shrinkage combination (DSC). Furthermore, our DSC approach significantly outperforms the naïve equal-weighted combination, solving the combination puzzle. The equal-weight shrinkage has greater effect in economic recessions, whereas the zero-weight shrinkage dominates in economic expansions. The DSC's superior performance over that of the naïve combination is observed in the application of forecasting macroeconomic indicators.</p>","PeriodicalId":54654,"journal":{"name":"Oxford Bulletin of Economics and Statistics","volume":"86 3","pages":"714-741"},"PeriodicalIF":2.5,"publicationDate":"2023-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138562830","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Fence off Black Swans: The Economics of Insurance for Vaccine Injury*","authors":"Ze Chen, Bingzheng Chen, Yu Mao","doi":"10.1111/obes.12585","DOIUrl":"10.1111/obes.12585","url":null,"abstract":"<p>Being injured during vaccination, although infrequent, can occur, and this necessitates understanding the consequences of vaccine injuries. In this study, we analyse the impact of the risk of vaccine injuries, particularly the economic importance of vaccine injury compensation insurance (VICI) programmes for the injured, which few studies have investigated. Specifically, we examine the role of VICI on individuals' vaccination decisions by integrating an imperfect vaccine into an epidemiological-economic model in the presence and absence of such an insurance mechanism. Our findings show that the risk of being injured is a non-negligible risk that lowers individuals' incentives to be vaccinated. The introduction of VICI largely alleviates decision-makers' concern about vaccine injury. Furthermore, our extended discussion shows the effect of insurance in encouraging vaccination persists under some more sophisticated scenarios.</p>","PeriodicalId":54654,"journal":{"name":"Oxford Bulletin of Economics and Statistics","volume":"86 5","pages":"995-1025"},"PeriodicalIF":1.5,"publicationDate":"2023-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138554831","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Opium Price Shocks and Prescription Opioids in the USA*","authors":"Claudio Deiana, Ludovica Giua, Roberto Nisticò","doi":"10.1111/obes.12584","DOIUrl":"10.1111/obes.12584","url":null,"abstract":"<p>We investigate the effect of international opium price shocks on the per capita dispensation of prescription opioids in the USA. Using quarterly county-level data for 2002q4–2016q4, three main results emerge. First, reductions in opium prices significantly increase the quantity of opioids prescribed, and more so in counties with a larger pre-existing market for pain relief, as captured by the incidence of mining sites. Second, the increase involves only natural and semi-synthetic, but not fully-synthetic, opioids, suggesting that the effect is moderated by the amount of raw material contained in the products. The impact is larger prior to 2010, when overdose deaths were more related to the use of legally prescribed opioids. Third, advertising expenses, stock prices and the profits of opioid producers increase following negative opium price shocks, suggesting an important role of supply-side economic incentives.</p>","PeriodicalId":54654,"journal":{"name":"Oxford Bulletin of Economics and Statistics","volume":"86 3","pages":"449-484"},"PeriodicalIF":2.5,"publicationDate":"2023-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/obes.12584","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138554756","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Partial Identification of Marginal Treatment Effects with Discrete Instruments and Misreported Treatment*","authors":"Santiago Acerenza","doi":"10.1111/obes.12581","DOIUrl":"10.1111/obes.12581","url":null,"abstract":"<p>This paper provides partial identification results for the marginal treatment effect (MTE) when the binary treatment variable is potentially misreported and the instrumental variable is discrete. Identification results are derived under smoothness assumptions. Bounds for both the case of misreported treatment and the case of no misreported treatment are derived. The identification results are illustrated by identifying the marginal treatment effects of food stamps on health.</p>","PeriodicalId":54654,"journal":{"name":"Oxford Bulletin of Economics and Statistics","volume":"86 1","pages":"74-100"},"PeriodicalIF":2.5,"publicationDate":"2023-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138554645","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Stephen Millard, Margarita Rubio, Alexandra Varadi
{"title":"The Macroprudential Toolkit: Effectiveness and Interactions","authors":"Stephen Millard, Margarita Rubio, Alexandra Varadi","doi":"10.1111/obes.12582","DOIUrl":"10.1111/obes.12582","url":null,"abstract":"<p>We use a DSGE model with financial frictions and with macroprudential limits on both banks and mortgage borrowers, in the form of capital requirements and maximum debt-service ratios. We then examine: (i) the impact of different combinations of macroprudential limits on key macroeconomic aggregates; (ii) their interaction with each other and with monetary policy; and (iii) their effects on the volatility of key macroeconomic variables and on welfare. We find that capital requirements on banks are the optimal tool when faced with a financial shock, as they nullify the effects of financial frictions and reduce the effects of the shock on the real economy. Instead, limits on mortgage debt-service ratios are optimal following a housing demand shock, as they disconnect the housing market from the real economy, reducing the volatility of inflation. Hence, no policy on its own is sufficient to deal with a wide range of shocks.</p>","PeriodicalId":54654,"journal":{"name":"Oxford Bulletin of Economics and Statistics","volume":"86 2","pages":"335-384"},"PeriodicalIF":2.5,"publicationDate":"2023-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/obes.12582","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138534686","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Identifying Politically Connected Firms: A Machine Learning Approach*","authors":"Vitezslav Titl, Deni Mazrekaj, Fritz Schiltz","doi":"10.1111/obes.12586","DOIUrl":"10.1111/obes.12586","url":null,"abstract":"<p>This article introduces machine learning techniques to identify politically connected firms. By assembling information from publicly available sources and the Orbis company database, we constructed a novel firm population dataset from Czechia in which various forms of political connections can be determined. The data about firms' connections are unique and comprehensive. They include political donations by the firm, having members of managerial boards who donated to a political party, and having members of boards who ran for political office. The results indicate that over 85% of firms with political connections can be accurately identified by the proposed algorithms. The model obtains this high accuracy by using only firm-level financial and industry indicators that are widely available in most countries. These findings suggest that machine learning algorithms could be used by public institutions to improve the identification of politically connected firms with potentially large conflicts of interest.</p>","PeriodicalId":54654,"journal":{"name":"Oxford Bulletin of Economics and Statistics","volume":"86 1","pages":"137-155"},"PeriodicalIF":2.5,"publicationDate":"2023-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/obes.12586","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138534687","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}