Inferential Theory for Factor Models of Large Dimensions

By Jushan Bai. 2003

1,790 citations recorded by the source corpus.

Subfields: Computational Theory and Mathematics, Economics and Econometrics, Statistics and Probability, Statistical Learning Theory

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

In Economics and Econometrics

In Statistics and Probability

In Statistical Learning Theory

Citations in the corpus

Machine Learning Methods That Economists Should Know About (2019, methodology): “This approach is analyzed in Bickel et al. [1998] and Van der Vaart [2000] for the general semiparametric case, and in Chernozhukov et al. [2017] for the average treatment effect case.”
Principal Components and Regularized Estimation of Factor Models (2017, methodology): “3 As as explained in Bai and Ng (2013), H̃NT will not, in general, be an identity matrix (2)The non-zero eigenvalues XX ′ and X X are the same.”
Sufficient Forecasting Using Factor Models (2015, methodology): “As an improvement to this procedure, Bair et al. (2006) used correlation screening to remove irrelevant predictors before performing the PCA. In a similar fashion, Bai and Ng (2008) applied boosting and employed thresholding rules to select ‘‘targeted predictors’’, and Stock and Watson (2012) used s”
Using Principal Component Analysis to Estimate a High Dimensional Factor Model with High-Frequency Data (2017, methodology): “Based on this model, Connor and Korajczyk (1993), Bai and Ng (2002), Amengual and Watson (2007), Onatski (2010) and Kapetanios (2010) propose statisticalmethodologies to determine the number of factors, while Bai (2003) provides tools to conduct statistical inference on the common factors and their ”
Understanding Systematic Risk: A High‐Frequency Approach (2020, neutral): “However, a shortcoming of the characteristic-sorted approach is that 2The general case of a static large-dimensional factor model is treated in Bai (2003) and Bai and Ng (2002).”
Eigenvalue Ratio Test for the Number of Factors (2013, methodology): “We use the same set of assumptions as in BN and Bai (2003) , but the”
EFFICIENT SEMIPARAMETRIC ESTIMATION OF THE FAMA-FRENCH MODEL AND EXTENSIONS (2012, methodology): “For each separate 12-month period July–June from 1970 to 2007, we find all securities which have complete Center for Research in Security Prices (CRSP) return records over this 12-month period and the previous 12-month period, and have two-digit Standard Industrial Classification code (from CRSP), m”
Statistical Inference for High-Dimensional Matrix-Variate Factor Models (2020, methodology): “While we are considering a static factor model (Bai, 2003) here, the information of ﬁrst moment, covariance and lag-h auto-covariance for h ≥ 1 can be aggregated to yield an even better performance, as long as E t is white noise.”
Non-Stationary Dynamic Factor Models for Large Datasets (2016, methodology): “So far, large-dimensional Dynamic Factor models have been studied mainly in a stationary setting, in which case the model can be consistently estimated either with principal components (Forni et al., 2000, 2005; Stock and Watson, 2002a; Bai and Ng, 2002; Bai, 2003), or with Maximum Likelihood by mea”
Systemic Risk and the Macroeconomy: An Empirical Evaluation (2016, positive): “12 This strips out variation in the target variable that is forecastable using its own history, following the forecasting literature such as Bai and Ng (2008b) and Stock and Watson (2012).”
Discretizing Unobserved Heterogeneity (2017, neutral): “Lastly, given parameter estimates α̂(k, j), σ̂(2), and θ̂, one can update the estimated partition of individuals using the full model’s structure, as in (8) and (9).”
Socioeconomic Status Measurement with Discrete Proxy Variables (2009, methodology): “, 2000), panel data (Bai, 1993; Reichlin, 2002), forecasting (Stock and Watson, 2002), simultaneous equations (Choi, 2002) and economics of education (Webster, 2001). Krelle (1997) gives a review of a number of methods aimed at estimation of unobservable variables, including PCA.”
Evaluating latent and observed factors in macroeconomics and finance (2006, background): “3 of Bai (2003), the last term is √ T ·Op(min[N, T ]−1) → 0 provided √ T/N → 0.”
Time-Varying Risk Premium in Large Cross-Sectional Equity Data Sets (2016, background): “3 in Bai and Ng (2002), except that we have an expectation w.”
A Testing Procedure for Determining the Number of Factors in Approximate Factor Models With Large Datasets (2010, methodology): “Then, under assumptions 1-5 and assumptions A-D of Bai (2003) the maximum eigenvalue of 1/T 2̂′2 is Op(1). We next discuss the subsampling methodology for estimating Ji. Subsampling was introduced informally by Mahalanobis (1946). Its properties were first discussed formally in Politis and Romano (1”
OPENING THE BLACK BOX: STRUCTURAL FACTOR MODELS WITH LARGE CROSS SECTIONS (2009, methodology): “Closely related models have been recently studied by Stock and Watson (2002a, 2002b), Bai and Ng (2002), and Bai (2003).”
Determining the Number of Factors from Empirical Distribution of Eigenvalues (2010, methodology): “In particular, the factors may follow I(1) processes as in Bai and Ng (2004). In addition, we do not require factors to be independent from the idiosyncratic terms, which are characterized by our next assumption.”