IECM 12.0 beta User Manual > Introduction to Uncertainty Analysis > Encoding Uncertainties as Probability Distributions >

Statistical Techniques

← Previous | Next →

Statistical estimation techniques involve estimating probability distributions from available data. The fit of data to a particular probability distribution function can be evaluated using various statistical tests. For example, the cumulative probability distribution of a set of data may be plotted on "probability" paper. If the data plot as a straight line, then the distribution is normal. Procedures for fitting probability distribution functions are discussed in many standard texts on probability and are not reviewed here.

Such procedures can be utilized to obtain distribution functions for many of the power plant parameters in the IECM when data are available for operating plants. In other cases, especially where data are limited, expert technical judgments may be necessary to develop appropriate distribution functions for model parameters. The emphasis of the discussion below is on the situations where statistical analysis alone may be insufficient.