34 cbluuisl dt errofbour satl ls ttar anndcahr de se sr or ol dr sb, ya st hr ee csoammme ei snsdueedr bayn dP ei nt e trhs ee ns a( 2m0e0y9e) a. r i n o r d e r t o Next, in order to investigate whether changing credit rating standards over time ha navdeLai un ai mn dpWa cat nogn (t2h0e1f9u)nadni ndgecsot ismt sa toef tahCe LfoOl laotwi si ns ug amnoc ed,ewl se: fo l l o w A l p ( 2 0 1 3 ) Where Rit denotes the credit rating of security i in issuance year t. αt is the intercept for year t, β is the vector of slope coefficients, and Zit is a latent variable that relates to Rit in the ranges between different partition points µi. Rit ranges from 1 to 21 as we have 21 rating categories in our sample. The matrix Xit denotes columns with explanatory variables. The variable definitions are given in Section 2.3.2. It nh eor redf oe rree dn ol ot gei ct omn oo md ei lcsa, l cl yo emf fei ca ineinntg fvuall, usei sn caer et hi en yuenai rt si nodf i ac al taot re nc to evfaf ri ci ai ebnl et a n d at is not in the same unit as Zit. Therefore, consistent with Alp (2013) and Liu and Wang (2019), we convert at into a rating notch, that is the average distance
RkJQdWJsaXNoZXIy MTk4NDMw