the estimated coefficient on ipo_size is also negative and statistically significant at the 1% level. We also see that, in the model, the adjusted R2 increase from 0.08 in model (2) to 0.49 in model (4), we include the variable reflecting change in stockholdings by insiders, CHANCE_BODS. as reported, the estimated coefficient is negative and statistically significant at the 10% level (t-statistics = -1.73). that is, the change in stockholdings by this group of investors one year after the offering is significantly negatively associated with the initial return or the underpricing of an IPO.