(Co-authored with Akifumi Ishihara)
Self-normalization if a tuning free inference method for time series that avoids long-run variance estimation. This talk will introduce the basic idea behind self-normalization and give intuition on when this method is applicable. We will also discuss the usage of self-normalization in two specific settings: change-point detection in the mean of high-dimensional time series and testing relevant hypotheses in functional time series.
Abstract. We propose confidence regions for the parameters of incomplete models
with exact coverage of the true parameter in finite samples. Our confidence region
inverts a test, which generalizes Monte Carlo tests to incomplete models. The test
statistic is a discrete analogue of a new optimal transport characterization of the
sharp identified region. Both test statistic and critical values rely on simulation
drawn from the distribution of latent variables and are computed using solutions
to discrete optimal transport, hence linear programming problems. We also pro-
pose a fast preliminary search in the parameter space with an alternative, more
conservative yet consistent test, based on a parameter free critical value.
要旨：Labor force participation of prime-aged males in the U.S. has secularly declined for a half century since 1970s, threatening the formation of partnership and fertility. I test a hypothesis that long-run climate change nudged their labor market exits, a significant fraction of whom are working outdoors under heightened exposure to hot days. Combining granular daily temperature data and labor force participation across U.S. Commuting Zones during 1970-2019, I find that accumulated exposure to hot day with mean temperature 80F accounts for 20-30% of increased non-participation of prime-aged males. Climate change significantly accounts for market exits of black males, historically agglomerated in the Southeast, where warming was severest.