We show that the higher-orders and their interactions of the common sparse linear factors can effectively subsume the factor zoo. To this extend, we propose a forward selection Fama-MacBeth procedure as a method to estimate a high-dimensional stochastic discount factor model, isolating the most relevant higher-order factors. Applying this approach to terms derived from six widely used factors (the Fama-French five-factor model and the momentum factor), we show that the resulting higher-order model with only a small number of selected higher-order terms significantly outperforms traditional benchmarks both in-sample and out-of-sample. Moreover, it effectively subsumes a majority of the factors from the extensive factor zoo, suggesting that the pricing power of most zoo factors is attributable to their exposure to higher-order terms of common linear factors.
We develop a theory of sectoral fluctuations driven by the propagation of demand shocks along supply chains with heterogeneous time-to-build production. We solve the model in closed form. Downstream producers respond directly to current demand. Upstream producers, due to time-to-build delays, respond to anticipated future demand. Consequently, hump-shaped demand shocks to downstream goods propagate and amplify along the supply chain, generating pronounced volatility in upstream sectors and creating the bullwhip effect. Empirically and quantitatively, we show that the bullwhip is significant across downstream sectors that are important for final consumption.
To safeguard economic and financial stability policymakers regularly take actions designed to increase resilience to systemic risks and curb speculative market behavior. To assess the effectiveness of such mitigation policies, we introduce a counterfactual approach tailored to accommodate the mildly explosive dynamics that occur during speculative bubbles. We derive asymptotics of the estimated treatment effect under a common factor structure that allows for explosive, I(1), and stationary factors, thereby having applicability to a wide range of prevailing economic conditions. An inferential procedure is proposed for the policy treatment effect that has asymptotic validity and demonstrates satisfactory finite sample performance. An empirical analysis examines the monetary policy of interest rate hikes implemented by the Reserve Bank of New Zealand, beginning in October 2021.This policy exerted a statistically significant cooling effect on all regional housing markets in New Zealand. Our findings show that this policy led to 20%-33% reductions in house prices in five out of six regions seven months after the enactment of the interest rate hike.
We study mechanism design when agents have private preferences and private information about a common payoff-relevant state. We show that standard message-driven mechanisms cannot implement socially efficient allocations when agents have multidimensional types, even under favorable conditions.
To overcome this limitation, we propose data-driven mechanisms that leverage additional post-allocation information, modeled as an estimator of the payoff-relevant state. Our data-driven mechanisms extend the classic Vickrey-Clarke-Groves class. We show that they achieve exact implementation in posterior equilibrium when the state is either fully revealed or the utility is affine in an unbiased estimator. We also show that they achieve approximate implementation with a consistent estimator, converging to exact implementation as the estimator converges, and present bounds on the convergence rate.
We demonstrate applications to digital advertising auctions and large language model (LLM)-based mechanisms, where user engagement naturally reveals relevant information.
Cointegrating rank selection is studied in a function space reduced rank regression where the data are time series of cross section curves. A semiparametric approach to rank selection is employed using information criteria suitably modified to take account of the function space context, extending the linear cointegrating model to accommodate cross section data under general forms of dependence. A parametric formulation is employed analogous to recent work on cross section curve autoregression and cointegrating regression. Consistent cointegrating rank estimation is developed by the use of information criteria methods that are extended to the curve time series environment. The asymptotic theory involves two parameter Gaussian processes that generalize the standard limit processes involved in cointegrating regressions with conventional multiple time series. Simulations provide evidence of the effectiveness of consistent rank selection by the BIC criterion and the tendency of AIC to overestimate order as it does in standard lag order selection in autoregression as well as in reduced rank regression with multiple time series.
We introduce a new methodology to detect and measure economic activity using geospatial data and apply it to steel production, a major industrial pollution source worldwide. Combining plant output data with geospatial data, such as ambient air pollutants, nighttime lights, and temperature, we train machine learning models to predict plant locations and output. We identify about 40% (70%) of plants missing from the training sample within a 1 km (5 km) radius and achieve R2 above 0.8 for output prediction at a 1 km grid and at the plant level, as well as for both regional and time series validations. Our approach can be adapted to other industries and regions, and used by policymakers and researchers to track and measure industrial activity in near real time.
Predictive regression models are often used to evaluate the predictive capability of economic fundamentals on bond and equity returns. Inferential procedures in these regressions typically employ parameter constancy or piecewise constancy in slope coefficients. Such formulations are prone to misspecification, more especially during periods of disturbance or evolution in prevailing economic and financial conditions, which can lead to size distortion and spurious evidence of predictability. To address these issues the present work proposes a semiparametric predictive regression model with mixed-root regressors and time-varying coefficients that allow for smooth evolution in the generating mechanism over time. For estimation and inference a novel variant of the self-generated instrument approach called Sieve-IVX is introduced, giving a robust approach to inference concerning time-varying predictability that is applicable irrespective of the degrees of persistence. Asymptotic theory of the Sieve-IVX approach is provided together with both pointwise and uniform inference procedures for testing predictability and model specification. Simulations show excellent performance characteristics of these statistics in finite samples. An empirical exercise is conducted to examine excess S&P 500 returns, applying Sieve-IVX regression coupled with pointwise and uniform tests to reveal evidence of time-varying patterns in the predictive capability of commonly used fundamental variables.
Recent literature suggests that both stock returns and economic growth are significantly higher under Democratic presidential administrations. This is a puzzle in that persistent differences in stock returns seem unlikely in efficient markets, and it is not obvious why Democrats should do better. Often these kinds of results go away upon further analysis or more data, and this appears to be true in the present case. In this paper the sample is extended to 28 administrations, fromWilson-1 through Biden. While the mean stock return under the Democrats is higher, none of the differences in means is significant at conventional significance levels. There is considerable variation in the mean return across administrations, which results in lack of significance. Similarly, while the mean output growth rate under the Democrats is larger, the difference is not significant. Again, there is considerable variation in output growth across administrations. Results are also presented with the nine administrations between Hayes and Taft added, a total of 37 administrations. While the added data are likely not as good, the conclusion is the same—no significant differences.
We analyze a nonlinear pricing model where the seller controls both product pricing (screening) and buyer information about their own values (persuasion).
We prove that the optimal mechanism always consists of finitely many signals and items, even with a continuum of buyer values. The seller optimally pools buyer values and reduces product variety to minimize informational rents.
We show that value pooling is optimal even for finite value distributions if their entropy exceeds a critical threshold. We also provide sufficient conditions under which the optimal menu restricts offering to a single item.
The recent artificial intelligence (AI) boom covers a period of rapid innovation and wide adoption of AI intelligence technologies across diverse industries. These developments have fueled an unprecedented frenzy in the Nasdaq, with AI-focused companies experiencing soaring stock prices that raise concerns about speculative bubbles and real-economy consequences. Against this background the present study investigates the formation of speculative bubbles in the Nasdaq stock market with a specific focus on the so-called ‘Magnificent Seven’ (Mag-7) individual stocks during the AI boom, spanning the period January 2017 to January 2025. We apply the real time PSY bubble detection methodology of Phillips et al. (2015a,b), while controlling for market and industry factors for individual stocks. Confidence intervals to assess the degree of speculative behavior in asset price dynamics are calculated using the near-unit root approach of Phillips (2023). The findings reveal the presence of speculative bubbles in the Nasdaq stock market and across all Mag-7 stocks. Nvidia and Microsoft experience the longest speculative periods over January 2017 – December 2021, while Nvidia and Tesla show the fastest rates of explosive behavior. Speculative bubbles persist in the market and in six of the seven stocks (excluding Apple) from December 2022 to January 2025. Near-unit-root inference indicates mildly explosive dynamics for Nvidia and Tesla (2017–2021) and local-to-unity near explosive behavior for all assets in both periods.