Parameter estimate for three-parameter kappa distribution using LHmoments approach

The method of higher-order L-moments (LH-moment) was proposed as a more robust alternative compared to classical L-moments to characterize extreme events. The new derivation will be done for Mielke-Johnson’s Kappa and Three-Parameters Kappa Type-II (K3D-II) distributions based on the LH-moments approach. The data of maximum monthly rainfall for Embong station in Terengganu were used as a case study. The analyses were conducted using the classical L-moments method with η=0 and LH-moments methods with η=1, η=2, η=3 and η=4 for a complete data series and upper parts of the distributions. The most suitable distributions were determined based on the Mean Absolute Deviation Index (MADI), Mean Square Deviation Index (MSDI), and Correlation (r). Also, L-moment and LH-moment ratio diagrams were used to represent visual proofs of the results. The analysis showed that LH-moments methods at a higher order of K3D-II distribution best fit the data of maximum monthly rainfalls for the Embong station for the upper parts of the distribution compared to L-moments. The results also proved that whenever η increases, LH-moments reflect more and more characteristics of the upper part of the distribution. This seems to suggest that LH-moments estimates for the upper part of the distribution events are superior to L-moments in fitting the data of maximum monthly rainfalls.

Regional frequency analysis of daily maximum rainfall in Haryana

This paper is sets-out for the regional frequency analysis of daily maximum rainfall from the 27 rain gauge stations in Haryana using L-moments. As the distribution of rainfall varies spatially in Haryana, the 27 rain gauge stations are grouped into three clusters namely, cluster C1, C2 and C3 using Ward’s clustering method and homogeneity of clusters was confirmed using L-moments-based Heterogeneity measure (H). Using goodness-of-fit measure ( DIST Z ) and L-moment ratios diagram, suitable regional frequency distributions were selected among five candidate distributions;Generalized Logistic (GLO), Generalized Extreme Value (GEV),Generalized Normal (GNO), Generalized Pareto (GPA), and Pearson Type-3 (PE3) for each cluster. Results showed that PE3 and GNO were good fitted regional distribution for the cluster C1 and GEV, PE3 and GNO fitted for cluster C2 while for cluster C3; GLO and GEV were good fitted regional distribution. To select a robust distribution among good fitted distributions accuracy measures calculated using Monte Carlo simulations for each cluster. The simulation result showed that PE3 was the best choice for quantile estimation for cluster C1. For cluster C2, PE3 was the best choicefor a large return period and GEV was best for a small return period. For cluster C3, GEV was the most suitable distribution for quantile estimation. Using these robust distributions rainfall quantiles were estimated at each rain gauge station from 2 to 100 year return periods. These estimated rainfall quantiles may be rough guideline for planning and designing hydraulic structures by policy makers and structural engineers.

Regional frequency analysis of daily maximum rainfall in Haryana

This paper is sets-out for the regional frequency analysis of daily maximum rainfall from the 27 rain gauge stations in Haryana using L-moments. As the distribution of rainfall varies spatially in Haryana, the 27 rain gauge stations are grouped into three clusters namely, cluster C1, C2 and C3 using Ward’s clustering method and homogeneity of clusters was confirmed using L-moments-based Heterogeneity measure (H). Using goodness-of-fit measure ( ) and L-moment ratios diagram, suitable regional frequency distributions were selected among five candidate distributions; Generalized Logistic (GLO), Generalized Extreme Value (GEV),Generalized Normal (GNO), Generalized Pareto (GPA), and Pearson Type-3 (PE3) for each cluster. Results showed that PE3 and GNO were good fitted regional distribution for the cluster C1 and GEV, PE3 and GNO fitted for cluster C2 while for cluster C3; GLO and GEV were good fitted regional distribution. To select a robust distribution among good fitted distributions accuracy measures calculated using Monte Carlo simulations for each cluster. The simulation result showed that PE3 was the best choice for quantile estimation for cluster C1. For cluster C2, PE3 was the best choicefor a large return period and GEV was best for a small return period. For cluster C3, GEV was the most suitable distribution for quantile estimation. Using these robust distributions rainfall quantiles were estimated at each rain gauge station from 2 to 100 year return periods. These estimated rainfall quantiles may be rough guideline for planning and designing hydraulic structures by policy makers and structural engineers.

L-Moments and Calibration-Based Estimators for Variance Parameter

The subject of variance estimation is one of the most important topics in statistics. It has been clarified by many different research studies due to its various applications in the human and natural sciences. Different variance estimators are built based on traditional moments that are especially influenced by the existence of extreme values. In this paper, with the presence of extreme values, we proposed some new calibration estimators for variance based on L-moments under double-stratified random sampling. A simulation study with COVID-19 data is performed to evaluate the efficiency of the proposed estimators. All results indicate that the proposed estimators are often superior and highly efficient compared to the existing traditional estimator.

Mean Estimators Using Robust Quantile Regression and L-Moments’ Characteristics for Complete and Partial Auxiliary Information

Ratio type regression estimator is a prevalent and readily implemented heuristic under simple random sampling (SRS) and two-stage sampling for the estimation of population. But this existing method is based on the ordinary least square (OLS) regression coefficient which is not an effective approach in the presence outliers in the data. In this article, we proposed a class of estimators firstly for complete auxiliary information and, later on, for partial auxiliary information for the presence of outliers in the data. To address this problem, initially we presented a distinct class of estimators by introducing the characteristics of L-moments in the existing estimators. Later on, quantile regression estimators are defined as more robust in the presence of outliers. These techniques empowered the proposed estimators to handle the problem of outliers. To prove the better performance of the proposed estimators, numerical studies are carried out using R language. To calculate the mean square error (MSE), hypothetical equations are expressed for adapted and proposed estimators. Percentage Relative Efficiencies (PRE) are compared to justify the proposed estimators.