Improved penalization for determining the number of factors in approximate factor models

By Lucia Alessi, Matteo Barigozzi, Marco Capasso. 2010

354 citations recorded by the source corpus.

Subfields: Economics and Econometrics, Finance, Statistics and Probability

How researchers see this paper

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

In Economics and Econometrics

In Finance

In Statistics and Probability

Citations in the corpus

Sufficient Forecasting Using Factor Models (2015, methodology): “There are many existing approaches to determining K in the literature, e.g., Bai and Ng (2002), Hallin and Liˇska (2007), Alessi et al. (2010).”
Large-dimensional factor modeling based on high-frequency observations (2019, neutral): “Their paper concentrates on the matrix-wise asymptotic 5There are many alternative methods, e.g. Hallin and Lisak (2007), Aumengual and Watson (2007), Alessi et al. (2010) or Kapetanious (2010), but in simulations they do not seem to outperform the above methods.”
Dynamic factor models with infinite-dimensional factor spaces: One-sided representations (2015, methodology): “Criteria to determine r consistently are given in Bai and Ng (2002) (see also Alessi et al. 2010).”
Dynamic Factor Models with Infinite-Dimensional Factor Space: Asymptotic Analysis (2017, background): “(i) The kernel function K is even, bounded, with support [−1, 1]; moreover, (1) for some κ > 0, limu→0 |K(u)− 1| = O(|u|κ), (2) ∫∞ −∞K 2(u)du <∞, (3) ∑ j∈Z sup|s−j|≤1 |K(jw)−K(sw)| = O(1) as w → 0; (ii) For some c1, c2 > 0, δ and δ such that 0 < δ < δ < 1 < δ(2κ+ 1), c1T δ ≤ BT ≤ c2T δ.”
Do Euro Area Countries Respond Asymmetrically to the Common Monetary Policy (2014, neutral): “Results for the criteria by Bai and Ng (2002), Hallin and Liška (2007), and Alessi et al. (2010), as well as robustness analysis for for q = {3, 5} and r = {9, 13}, are available in a complementary appendix to this paper.”
Using Principal Component Analysis to Estimate a High Dimensional Factor Model with High-Frequency Data (2017, methodology): “Alternative methods to determine the number of factors include Hallin and Lǐska (2007), Amengual and Watson (2007), Alessi et al. (2010), Kapetanios (2010), and Onatski (2010).”
A Randomized Sequential Procedure to Determine the Number of Factors (2018, methodology): “The estimator developed by Alessi, Barigozzi, and Capasso (2010) performs also very well, as long as N is sufficiently large (at least for the case of cross-sectional dependence).”
Generalized dynamic factor models and volatilities: estimation and forecasting (2017, neutral): “…(as in Stock and Watson, 2002, for instance); in a second step, an AR-GARCH model is adjusted to each estimated principal component.16 The numberr of static factors is selected by means of the criteria of Bai and Ng (2002) or Alessi et al. (2010); both criteria identify three static common factors.”
Generalized dynamic factor models and volatilities: recovering the market volatility shocks (2016, neutral): “This is confirmed by the second part of Table 1, where, along with the MSEs defined in (4.1) for the static approach, we also report the average (over all replications) number of factors identified by the Alessi et al. (2010) criterion (r for the levels; r s and r w for the volatility panels).”
An Overview on the Estimation of Large Covariance and Precision Matrices (2015, neutral): “ is the number of factors and practically unknown. There has been a large literature on determining the number of factors and many consistent estimators have been proposed, such as Bai and Ng (2002); Alessi et al. (2010); Hallin and Liska (2007), and Ahn and Horenstein (2013). In addition, numerical”
A diagnostic criterion for approximate factor structure (2016, methodology): “The constant is selected as in Alessi et al. (2010) (see also Hallin and Liška (2007) in the general dynamic factor model).”
Simultaneous multiple change-point and factor analysis for high-dimensional time series (2016, background): “However, the true number of factors r is typically unknown and, since the seminal paper by Bai andNg (2002), estimation of the number of factors has been one of the most researched problems in the factor model literature see also Alessi et al. (2010), Onatski (2010) and Ahn and Horenstein (2013).”
Bi-cross-validation for factor analysis (2015, methodology): “Some of the most popular methods to estimate the number of factors under the above scenario are based on the information criteria developed by Bai and Ng [5], with later improvements in [2].”
Efficient Estimation of Approximate Factor Models (2012, methodology): “In practice, r can be estimated from the data, and there has been a large literature addressing its consistent estimation, e.g., Bai and Ng (2002), Kapetanios (2010), Onatski (2010), Alessi et al. (2010), Hallin and Lǐska (2007), Lam and Yao (2012), among others.”
Large Covariance Estimation by Thresholding Principal Orthogonal Complements (2011, neutral): “Other recent criteria are proposed by Kapetanios (2010), Onatski (2010), Alessi et al. (2010), etc.”
Non-Stationary Dynamic Factor Models for Large Datasets (2016, methodology): “We find evidence of r = 7 common factors as suggested both by the criteria in Alessi et al. (2010) and in Bai and Ng (2002), and of q = 3 common shocks as given by the criterion in Hallin and Liška (2007).”
Factor Models with Local Factors — Determining the Number of Relevant Factors (2021, methodology): “To address the problem that the estimated number of factors is sensitive to the choice of cutoff under existing thresholding estimators, Alessi et al. (2010) suggest varying this threshold and ex10All variables have been transformed to achieve stationarity and a small number of outliers were removed”
PROJECTED PRINCIPAL COMPONENT ANALYSIS IN FACTOR MODELS (2014, positive): “In principle many consistent estimators of K can be employed, e.g., Bai and Ng (2002), Alessi et al. (2010), Breitung and Pigorsch (2009), Hallin and Liˇska (2007).”