I am conducting a research on the cost of equity. I followed the research model from a past literature but since I'm unable to obtain the forecasting EPS for PEG ratio, I'm using other PEG formula (PER/Earnings growth). My data contains many negative PEG and there are many outliers. How to deal with the negative PEG to get the correct result? Is it possible to put 0 to all the negative PEG? Referring a journal article would be great!