A wavelet method for panel models with jump discontinuities in the parameters

While a substantial literature on structural break change point analysis exists for univariate time series, research large panel data models has not been as extensive. In this paper, novel method estimating with multiple changes is proposed. The breaks are allowed to occur at unknown points in and may affect the multivariate slope parameters individually. Our adapts Haar wavelets structure of observed variables order detect consistently. We also develop methods address endogenous regressors within our modeling framework. asymptotic property estimator established. application, we examine impact algorithmic trading standard measures market quality such liquidity volatility over period that covers financial meltdown began 2007. able jumps regression automatically without using ad-hoc subsample selection criteria.