Confidence Interval Methods

The confidence interval is a measure of the reliability of the estimated parameters. In the current implementation only the Fisher’s Information method is available and it is automatically executed after the estimation has completed. As illustrated in the estimation results section DEDiscover will compute the p-value, t-statistics, confidence interval upper and lower bounds, as well as the Standard error for each parameter.

Fisher’s Information Method

The Fisher’s information method makes some very strong assumptions which are unlikely to hold for most models, and so should not be considered to be very reliable. However it is not computationally demanding.

The only parameter in the Fisher’s information is the alpha parameter, which is the significance level.