Outliers in Spectral Time Lag-Selected Gamma Ray Bursts

It is possible that astrophysical samples are polluted by some outliers, which might belong to a different sub-class. By removing the underlying statistical features may be revealed. A more reliable correlation can used as standard candle relationship for cosmological study. We present outlier searching gamma ray bursts with Partitioning Around Medoids (PAM) method. In this work, we choose three parameters from sample, all of them having rest-frame spectral time lag (τlag,i). most cases, outliers GRBs 980425B and 030528A. Linear regression carried out sample without outliers. Some have passed hypothesis testing, while others not. However, even not very significant. More parameter combinations should considered in future work.