Forecasts play a central role in decision-making under uncertainty. After a brief review of the general issues, this article considers ways of using high-dimensional data in forecasting. We consider selecting variables from a known active set, known knowns, using Lasso and One Covariate at a time Multiple Testing, and approximating unobserved latent factors, known unknowns, by various means. This combines both sparse and dense approaches to forecasting. We demonstrate the various issues involved in variable selection in a high-dimensional setting with an application to forecasting UK inflation at different horizons over the period 2020q1–2023q1. This application shows both the power of parsimonious models and the importance of allowing for global variables.

Researchers of time series cross-sectional data regularly face the change-point problem, which requires them to discern between significant parametric shifts that can be deemed structural changes and minor parametric shifts that must be considered noise. In this paper, we develop a general Bayesian method for change-point detection in high-dimensional data and present its application in the context of the fixed-effect model. Our proposed method, hidden Markov Bayesian bridge model, jointly estimates high-dimensional regime-specific parameters and hidden regime transitions in a unified way. We apply our method to Alvarez, Garrett, and Lange’s (1991, American Political Science Review 85, 539–556) study of the relationship between government partisanship and economic growth and Allee and Scalera’s (2012, International Organization 66, 243–276) study of membership effects in international organizations. In both applications, we found that the proposed method successfully identify substantively meaningful temporal heterogeneity in parameters of regression models.

In general, the clustering problem is NP-hard, and global optimality cannot be established for non-trivial instances. For high-dimensional data, distance-based methods for clustering or classification face an additional difficulty, the unreliability of distances in very high-dimensional spaces. We propose a probabilistic, distance-based, iterative method for clustering data in very high-dimensional space, using the ℓ1-metric that is less sensitive to high dimensionality than the Euclidean distance. For K clusters in ℝn, the problem decomposes to K problems coupled by probabilities, and an iteration reduces to finding Kn weighted medians of points on a line. The complexity of the algorithm is linear in the dimension of the data space, and its performance was observed to improve significantly as the dimension increases.