Integral inference method for scatter and probability

An integral inference method for scatter and probability of normal distribution with zero mean is presented. By combining the results of simulation and the experiment data, the method can totally infer the experiment data of normal distributions in multi-state. Furthermore, it can establish the variance and the probability estimators of normal distribution, and give their confidence limits and intervals in each state. Compared with traditional methods, this method not only has higher precision but also solves the problem of scatter and probability assessment with only one experiment datum of normal distribution in a state. A test method for simulation results of variances in multi-state with very small sample is also presented.