Many time series exhibit “long memory”: Their autocorrelation function decays slowly with lag. This behavior has traditionally been modeled via unit roots or fractional Brownian motion and explained via aggregation of heterogeneous processes, nonlinearity, learning dynamics, regime switching, or structural breaks. This paper identifies a different and complementary mechanism for long‐memory generation by showing that it can naturally arise when a large number of simple linear homogeneous economic subsystems with short memory are interconnected to form a network such that the outputs of the subsystems are fed into the inputs of others. This networking picture yields a type of aggregation that is not merely additive, resulting in a collective behavior that is richer than that of individual subsystems. Interestingly, the long‐memory behavior is found to be almost entirely determined by the geometry of the network, while being relatively insensitive to the specific behavior of individual agents.

We introduce firm and worker heterogeneity into a model of innovation†driven endogenous growth. Individuals who differ in ability sort into either a research activity or a manufacturing sector. Research projects generate new varieties of a differentiated product. Projects differ in quality and the resulting technologies differ in productivity. In both sectors, there is a complementarity between firm quality and worker ability. We study the co†determination of growth and income inequality in both the closed and open economy, as well as the spillover effects of policy in one country to outcomes in others.

We develop inference methods about long†run comovement of two time series. The parameters of interest are defined in terms of population second moments of low†frequency transformations (“low†pass†filtered versions) of the data. We numerically determine confidence sets that control coverage over a wide range of potential bivariate persistence patterns, which include arbitrary linear combinations of I(0), I(1), near unit roots, and fractionally integrated processes. In an application to U.S. economic data, we quantify the long†run covariability of a variety of series, such as those giving rise to balanced growth, nominal exchange rates and relative nominal prices, the unemployment rate and inflation, money growth and inflation, earnings and stock prices, etc.

It is generally difficult to know whether the parameters in nonlinear econometric models are point‐identified. We provide computationally attractive procedures to construct confidence sets (CSs) for identified sets of the full parameter vector and of subvectors in models defined through a likelihood or a vector of moment equalities or inequalities. The CSs are based on level sets of “optimal” criterion functions (such as likelihoods, optimally‐weighted or continuously‐updated GMM criterions). The level sets are constructed using cutoffs that are computed via Monte Carlo (MC) simulations from the quasi‐posterior distribution of the criterion. We establish new Bernstein–von Mises (or Bayesian Wilks) type theorems for the quasi‐posterior distributions of the quasi‐likelihood ratio (QLR) and profile QLR in partially‐identified models. These results imply that our MC CSs have exact asymptotic frequentist coverage for identified sets of full parameters and of subvectors in partially‐identified regular models, and have valid but potentially conservative coverage in models whose local tangent spaces are convex cones. Further, our MC CSs for identified sets of subvectors are shown to have exact asymptotic coverage in models with singularities. We provide local power properties and uniform validity of our CSs over classes of DGPs that include point‐ and partially‐identified models. Finally, we present two simulation experiments and two empirical examples: an airline entry game and a model of trade flows.

This paper provides an alternative approach to penalized regression for model selection in the context of high‐dimensional linear regressions where the number of covariates is large, often much larger than the number of available observations. We consider the statistical significance of individual covariates one at a time, while taking full account of the multiple testing nature of the inferential problem involved. We refer to the proposed method as One Covariate at a Time Multiple Testing (OCMT) procedure, and use ideas from the multiple testing literature to control the probability of selecting the approximating model, the false positive rate, and the false discovery rate. OCMT is easy to interpret, relates to classical statistical analysis, is valid under general assumptions, is faster to compute, and performs well in small samples. The usefulness of OCMT is also illustrated by an empirical application to forecasting U.S. output growth and inflation.