A discrete‐choice model for large heterogeneous panels with interactive fixed effects with an application to the determinants of corporate bond issuance

Abstract

Summary What is the effect of funding costs on the conditional probability of issuing a corporate bond? We study this question in a novel dataset covering 5610 issuances by US firms over the period from 1990 to 2014. Identification of this effect is complicated because of unobserved, common shocks such as the global financial crisis. To account for these shocks, we extend the common correlated effects estimator to settings where outcomes are discrete. Both the asymptotic properties and the small‐sample behavior of this estimator are documented. We find that for non‐financial firms yields are negatively related to bond issuance but that the effect is larger in the pre‐crisis period.

Common factors and common breaks in panels: An empirical investigation

“…In column (4) of Table 3, using the estimation procedure in BFK (2016), the estimated common break 2004 splits b and k in two regimes of 1997-2004 and 2005-2015. 4 The CCEMG-b estimates of b and k deviate moderately from their CCEMG counterparts in column (3) of Table 3 in di¤erent directions.…”

“…In this case, the CCEMG-IV-b estimate of k becomes 0.765 and signi…cant over the period 2008-2015, but the coe¢ cient of b is still insigni…cant. Comparison 4 There are empirical studies that provide evidence why 2004 could be a break point. Lu and Xiang (2016) …nd the overall slowdown of economic e¢ ciency in China after 2003 and link it to a clear policy turning point in the regional allocation of economic resources.…”

Incidental Parameters Problem: The Case of Gompertz Model

“…A few of several other works in both static and dynamic panel data models include Greene [10]; Moreira [11]; Hahn and Kuersteiner [12] and Moon, Perron and Philips [13]. More recent works include Femendez-Val and Weidner [14]; Moon and Wedner [15]; Boneva and Linton [16] and Juodis [17]. This article studies the order of bias of fixed effects gompertz model.…”

Large-scale generalized linear longitudinal data models with grouped patterns of unobserved heterogeneity

“…Studies on clustering longitudinal data with unobserved heterogeneity in the context of an exponential family of distributions are scant. Many previous studies on binary response longitudinal data with unob-served effects (for, e.g., Fernández-Val and Weidner (2016), Charbonneau (2017), Moon et al (2018), Boneva and Linton (2017), Ando et al (2021), Chen et al (2021)) did not accommodate the unobserved group membership structure. Additionally, many studies have considered group-specific regression coefficients, and they all assumed that the regression coefficients are homogeneous across units within groups but can differ across groups.…”