Economics

We propose a novel structural estimation framework in which we train a surrogate of an economic model with deep neural networks. Our methodology alleviates the curse of dimensionality and speeds up the evaluation and parameter estimation by orders of magnitudes, which significantly enhances one's ability to conduct analyses that require frequent parameter re-estimation. As an empirical application, we compare two popular option pricing models (the Heston and the Bates model with double-exponential jumps) against a non-parametric random forest model. We document that: a) the Bates model produces better out-of-sample pricing on average, but both structural models fail to outperform random forest for large areas of the volatility surface; b) random forest is more competitive at short horizons (e.g., 1-day), for short-dated options (with less than 7 days to maturity), and on days with poor liquidity; c) both structural models outperform random forest in out-of-sample delta hedging; d) the Heston model's relative performance has deteriorated significantly after the 2008 financial crisis.

We develop inference procedures robust to general forms of weak dependence. The procedures use test statistics constructed by resampling data in a manner that does not depend on the unknown correlation structure of the data. We prove that the statistics are asymptotically normal under the weak requirement that the target parameter can be consistently estimated at the parametric rate. This holds for regular estimators under many well-known forms of weak dependence and justifies the claim of dependence-robustness. We consider applications to settings with unknown or complicated forms of dependence, with various forms of network dependence as leading examples. We develop tests for both moment equalities and inequalities.

Social scientists are often interested in estimating causal effects in settings where all units in the population are observed (e.g. all 50 US states). Design-based approaches, which view the treatment as the random object of interest, may be more appealing than standard sampling-based approaches in such contexts. This paper develops a design-based theory of uncertainty suitable for quasi-experimental settings, in which the researcher estimates the treatment effect as if treatment was randomly assigned, but in reality treatment probabilities may depend in unknown ways on the potential outcomes. We first study the properties of the simple difference-in-means (SDIM) estimator. The SDIM is unbiased for a finite-population design-based analog to the average treatment effect on the treated (ATT) if treatment probabilities are uncorrelated with the potential outcomes in a finite population sense. We further derive expressions for the variance of the SDIM estimator and a central limit theorem under sequences of finite populations with growing sample size. We then show how our results can be applied to analyze the distribution and estimand of difference-in-differences (DiD) and two-stage least squares (2SLS) from a design-based perspective when treatment is not completely randomly assigned.

This paper proposes a logistic undirected network formation model which allows for assortative matching on observed individual characteristics and the presence of edge-wise fixed effects. We model the coefficients of observed characteristics to have a latent community structure and the edge-wise fixed effects to be of low rank. We propose a multi-step estimation procedure involving nuclear norm regularization, sample splitting, iterative logistic regression and spectral clustering to detect the latent communities. We show that the latent communities can be exactly recovered when the expected degree of the network is of order log n or higher, where n is the number of nodes in the network. The finite sample performance of the new estimation and inference methods is illustrated through both simulated and real datasets.

This paper investigates the added benefit of internet search data in the form of Google Trends for nowcasting real U.S. GDP growth in real time through the lens of the mixed frequency augmented Bayesian Structural Time Series model (BSTS) of Scott and Varian (2014). We show that a large dimensional set of search terms are able to improve nowcasts before other macro data becomes available early on the quarter. Search terms with high inclusion probability have negative correlation with GDP growth, which we reason to stem from them signalling special attention likely due to expected large troughs. We further offer several improvements on the priors: we allow to shrink state variances to zero to avoid overfitting states, extend the SSVS prior to the more flexible normal-inverse-gamma prior of Ishwaran et al. (2005) which stays agnostic about the underlying model size, as well as adapt the horseshoe prior of Carvalho et al. (2010) to the BSTS. The application to nowcasting GDP growth as well as a simulation study show that the horseshoe prior BSTS improves markedly over the SSVS and the original BSTS model, with largest gains to be expected in dense data-generating-processes.

We consider the estimation of the effect of a treatment, using panel data where groups of units are exposed to different doses of the treatment at different times. We consider two sets of parameters of interest. The first are the average effects of having changed treatment for the first time ℓ periods ago. Those parameters generalize the average effect of having started receiving the treatment ℓ periods ago that has often been estimated in applications with a binary treatment and staggered adoption. We also consider cost-benefit ratios a planner may use to compare the treatments actually assigned to a counterfactual \textit{status quo} scenario where groups receive their period-one treatment throughout the panel. We show that under common trends conditions, all these parameters are unbiasedly estimated by weighted sums of differences-in-differences. Our estimators are valid if the treatment effect is heterogeneous, contrary to the commonly-used dynamic two-way fixed effects regressions. We also propose estimators of a dynamic linear model, with group-specific but time-invariant effects of the current and lagged treatments, which may be used to evaluate ex-ante the effect of future policies. In an application, we find that our estimators differ substantially from those obtained with dynamic two-way fixed effects regressions.

We propose a new semiparametric approach for modelling nonlinear univariate diffusions, where the observed process is a nonparametric transformation of an underlying parametric diffusion (UPD). This modelling strategy yields a general class of semiparametric Markov diffusion models with parametric dynamic copulas and nonparametric marginal distributions. We provide primitive conditions for the identification of the UPD parameters together with the unknown transformations from discrete samples. Likelihood-based estimators of both parametric and nonparametric components are developed and we analyze the asymptotic properties of these. Kernel-based drift and diffusion estimators are also proposed and shown to be normally distributed in large samples. A simulation study investigates the finite sample performance of our estimators in the context of modelling US short-term interest rates. We also present a simple application of the proposed method for modelling the CBOE volatility index data.

We propose a methodology for constructing confidence regions with partially identified models of general form. The region is obtained by inverting a test of internal consistency of the econometric structure. We develop a dilation bootstrap methodology to deal with sampling uncertainty without reference to the hypothesized economic structure. It requires bootstrapping the quantile process for univariate data and a novel generalization of the latter to higher dimensions. Once the dilation is chosen to control the confidence level, the unknown true distribution of the observed data can be replaced by the known empirical distribution and confidence regions can then be obtained as in Galichon and Henry (2011) and Beresteanu, Molchanov and Molinari (2011).

This paper aims to decompose a large dimensional vector autoregessive (VAR) model into two components, the first one being generated by a small-scale VAR and the second one being a white noise sequence. Hence, a reduced number of common factors generates the entire dynamics of the large system through a VAR structure. This modelling extends the common feature approach to high dimensional systems, and it differs from the dynamic factor models in which the idiosyncratic components can also embed a dynamic pattern. We show the conditions under which this decomposition exists, and we provide statistical tools to detect its presence in the data and to estimate the parameters of the underlying small-scale VAR model. We evaluate the practical value of the proposed methodology by simulations as well as by empirical applications on both economic and financial time series.

We propose a novel approach for causal mediation analysis based on changes-in-changes assumptions restricting unobserved heterogeneity over time. This allows disentangling the causal effect of a binary treatment on a continuous outcome into an indirect effect operating through a binary intermediate variable (called mediator) and a direct effect running via other causal mechanisms. We identify average and quantile direct and indirect effects for various subgroups under the condition that the outcome is monotonic in the unobserved heterogeneity and that the distribution of the latter does not change over time conditional on the treatment and the mediator. We also provide a simulation study and an empirical application to the Jobs II programme.