Search Results
Working Paper
Predicting Benchmarked US State Employment Data in Real Time
US payroll employment data come from a survey and are subject to revisions. While revisions are generally small at the national level, they can be large enough at the state level to alter assessments of current economic conditions. Users must therefore exercise caution in interpreting state employment data until they are “benchmarked” against administrative data 5–16 months after the reference period. This paper develops a state-space model that predicts benchmarked state employment data in real time. The model has two distinct features: 1) an explicit model of the data revision process ...
Working Paper
A Note on the Finite Sample Bias in Time Series Cross-Validation
It is well known that model selection via cross validation can be biased for time series models. However, many researchers have argued that this bias does not apply when using cross-validation with vector autoregressions (VAR) or with time series models whose errors follow a martingale-like structure. I show that even under these circumstances, performing cross-validation on time series data will still generate bias in general.
Working Paper
Decomposition of feedback between time series in a bivariate error-correction model
This paper adapts Geweke's [1982] method of decomposing the feedback between time series by frequency to the case of 1(1) time series generated by a bivariate error-correction model. The method is applied to long-run data on US and UK price levels with the finding that most of the feedback between the two time series occurs at very low frequencies.
Working Paper
A Note on the Finite Sample Bias in Time Series Cross-Validation
It is well known that model selection via cross validation can be biased for time series models. However, many researchers have argued that this bias does not apply when using cross-validation with vector autoregressions (VAR) or with time series models whose errors follow a martingale-like structure. I show that even under these circumstances, performing cross-validation on time series data will still generate bias in general.
Working Paper
xtcipsunb: The CIPS Panel Unit Root Test for Unbalanced Panel Data
We develop and demonstrate the command xtcipsunb, which implements the cross-sectionally augmented panel unit root test (CIPS) from Pesaran (2007) and Pesaran, Smith and Yamagata (2013) for unbalanced panels. Several modifications relative to the existing Stata implementations of CIPS test are necessitated by the unbalanced panel data setting, including computing critical values through simulations for a given panel composition. We provide users the ability to specify a minimum number of cross-section units for the computation of cross-section averages, a minimum number of time periods for ...