Abstract:
The objective of this research is to estimate the probability density function when observations are mixed from a chi-square distribution and an exponential distribution which have different parameters. We compare four estimation methods; First Ordinary Moment, Secondary Ordinary Moment, Secondary Moment about the mean and Third Moment about the mean for finding probability density function by considering Mean Square Error : MSE and Mean Integrated Square Error : MISE. The second moment about the mean is suitable when the data are from chi-square distribution with high skewness, otherwise, the second ordinary moment is preferable for the cases of low skewness. For the case of hyper-exponential distribution, the first ordinary moment is suitable for high skewness, otherwise, the second moment about the mean is preferable for the cases of low skewness.