Engineered Software

Parameter Estimation


Techniques

Cumulative Distribution Function

Cumulative Hazard Function

Weibull Distribution

Normal Distribution

Lognormal Distribution

Exponential Distribution

Exam

Engineered Software Home Page

There are many methods available for parameter estimation. In reliability engineering, the most popular methods are:

  • Maximum likelihood
  • Hazard plotting
  • Probability plotting
  • Moment estimation

It is desirable for a parameter estimator to have the following properties:

  1. Lack of bias—if the expected value of the estimator is equal to the true value of the parameter, it is said to be unbiased.

  2. Minimum variance—the smaller the variance of the estimate, the smaller the sample size required to obtain the level of accuracy desired, and the more efficient the estimator. The most efficient estimator is the estimator with minimum variance.

  3. Consistency—as the sample size is increased, the value of the estimated parameter becomes closer to the true value of the parameter.

  4. Sufficiency—the estimator uses all information available in the data set.

Parameter estimation techniques are tedious, and computers are often employed to aid in data analysis. This section provides a detailed explanation for each parameter estimation technique as well as a demonstration using The Reliability & Maintenance Analyst software package.

BackHomeNext