| Author | Chien, Kian-cheng |
| Call Number | AIT Thesis no. CS-87-4 |
| Subject(s) | Distribution, Theory of (Functional analysis)
|
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering, School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Abstract | There have been many studies on the methods of parameter
estimation for the Generalized Gamma Distribution, however the
maxi mum likelihood method with censoring and the Sundry Average
Method have not been considered. The criteria used for the
assessment of the performance of these methods were the relative
error or the percent standard error of the parameter estimators.
In this study, an attemp was first made to derive formulas
for parameter estimation and variances/covariances of the
parameter estimators for the above mentioned methods. The
criteria used in this study are the relative bias, the root mean
square error CRMSE, and the sum of the absolute value of the
relative bias and RMSE denoted by RMSBI of the T-year event.
Related computer programs were developed with good documentation.
From comparison between methods of parameter estimation for
complete samples, the method based upon the first moment of the
GGD variable and the mean and variance of its logarithm has the
best performance in terms of the RMSE. The method based upon the
arithmetic, geometric and harmonic mean of the GGD variables has
the best performance with respect to the relative bias.
From the investigation of the effect of censoring, it was
found that censoring Cleft, right and double> can reduce the
relative bias. Moreover, censoring to the right was found to deal
with C large) outliers very well.
The Kolmogorov- Smirnov and Chi - Square tests showed that the
GGD can fit many annual flood data sets.
Finally, having been tested in many cases by both simulated
and actual data, the programs developed in the study can be
readily used for estimating the parameters and the T-year events. |
| Year | 1987 |
| Type | Thesis |
| School | School of Engineering and Technology (SET) |
| Department | Other Field of Studies (No Department) |
| Academic Program/FoS | Computer Science (CS) |
| Chairperson(s) | Huynh Ngoc Phien; |
| Examination Committee(s) | Venkatachalam, Paruvachi A. ; Sadananda, Ramakoti |
| Scholarship Donor(s) | The Government of R. 0 . C. ; |
| Degree | Thesis (M.Eng.) - Asian Institute of Technology, 1987 |