The purpose of asset write-offs by a firm is to provide an accurate valuation of thefirm and to reveal its true business performance from the perspective of economic conditions.However, the decision to write-off assets might be manipulated by the manager of the firm andthus misguide the public to an incorrect firm value. The aim of this study is to provide quantitativeprediction models for asset write-offs based on both firms’ financial and managerial incentivefactors. The prediction is achieved in two stages, where the first stage conducts a binary predictionof the occurrence of asset write-offs by a firm, while the second stage predicts the magnitudeof such asset write-offs if they took place. The prediction models are constructed by supportvector machine (SVM) and logistic regression for the binary decision of asset write-offs,and support vector regression (SVR) and linear regression for the write-off magnitude. The performancesof different models are compared in terms of various criteria. Moreover, the baggingapproach is used to reduce the variance in samples to improve prediction performance. Computationalresults from empirical data show the prediction performances of SVM/SVR are moderatelysuperior to their counterpart logit/linear models. Moreover, the prediction accuracy varieswith the distinctive types of asset write-offs.