This paper develops an asymptotic theory for estimated change-points in linear and nonlinear time series models. Based on a measurable objective function, it is shown that the estimated change-point converges weakly to the location of the maxima of a double-sided random walk and other estimated parameters are asymptotically normal. When the magnitude d of changed parameters is small, it is shown that the limiting distribution can be approximated by the known distribution as in Yao (1987, Annals of Statistics 15, 1321-1328. This provides a channel to connect our results with those in Picard (1985, Advances in Applied Probability 17, 841-867) and Bai, Lumsdaine, and Stock (1998, Review of Economic Studies 65, 395-432), where the magnitude of changed parameters depends on the sample size n and tends to zero as n → ∞. The theory is applied for the self-weighted QMLE and the local QMLE of change-points in ARMA-GARCH/IGARCH models. A simulation study is carried out to evaluate the performance of these estimators in the finite sample. 1 2 SHIQING LING and Eickmeier (2011) and Han and Inoue (2014) studied some tests for structural breaks in dynamic factor models. Ling (2007a) developed an asymptotic theory of the Quandt-type tests for linear and nonlinear time series models. Aue, Hörmann, Horváth, and Reimherr (2009) studied the break detection in the covariance structure of multivariate time series models. Shao and Zhang (2010) studied Quandt-type test for the change of mean in time series. This type of tests was further developed by Hidalgo and Seo (2013) under a larger framework. Empirically, we want to know not only that structural change exists, but also the location of change-point.The first paper on the estimation of change-points is by Hinkley (1970), in which he investigated the maximum likelihood estimator (MLE) of the change-points in a sequence of i.i.d. random variables and proved that the estimated change-point converges in distribution to the location of the maxima of a double-sided random walk. Under the normality assumption, he showed that the limiting distribution can be tabulated by a numerical method. Hinkley and Hinkley (1970) used a similar method to investigate the binomial random variables and showed that the limiting distribution has a computable form. However, for the nonnormal or nonbinomial cases, their results cannot be used as statistical inference for the change point. When the magnitude d of changed parameters is small, Yao (1987) showed that Hinkley's (1970) limiting distribution can be approximated by a very nice distribution. Ritov (1990) studied the asymptotic efficient estimation of the change-point. Dümbgen (1991) investigated the nonparametric method for change-point estimators. Bai (1995) studied a structure-changed regression model with a fixed d and showed that the estimated change-point converges in distribution to the location of the maxima of a double-sided random walk. Qu and Perron (2007) investigated estimating and testing structural changes in multivariate regressions. H...