1999
DOI: 10.1155/s1173912699000012
|View full text |Cite
|
Sign up to set email alerts
|

Robustness of the sample correlation - the bivariate lognormal case

Abstract: Abstract. The sample correlation coefficient R is almost universally used to estimate the population correlation coefficient p. If the pair (X, Y)has a bivariate normal distribution, this would not cause any trouble. However, if the marginals are nonnormal, particularly if they have high skewness and kurtosis, the estimated value from a sample may be quite different from the population correlation coefficient p. The bivariate lognormal is chosen as our case study for this robustness study. Two approaches are u… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

2
15
0

Year Published

2002
2002
2022
2022

Publication Types

Select...
7
2
1

Relationship

0
10

Authors

Journals

citations
Cited by 22 publications
(17 citation statements)
references
References 3 publications
(4 reference statements)
2
15
0
Order By: Relevance
“…The second part of Table 1 lists ASV of all correlations under the lognormal distribution with σ = and σ = . In this case, the Pearson correlation has extremely large asymptotic variances, the result agreeing well with [19,23]. The asymptotic variance ofrp involves the fourth moment and is given by Witting and Müller-Funk ( [40]) as follows.…”
Section: Asymptotic Relative E Ciencysupporting
confidence: 53%
“…The second part of Table 1 lists ASV of all correlations under the lognormal distribution with σ = and σ = . In this case, the Pearson correlation has extremely large asymptotic variances, the result agreeing well with [19,23]. The asymptotic variance ofrp involves the fourth moment and is given by Witting and Müller-Funk ( [40]) as follows.…”
Section: Asymptotic Relative E Ciencysupporting
confidence: 53%
“…Numerous authors reviewed by Johnson et al (1995) and Lai et al (1999) have derived expressions for the first four moments of r in terms of the cumulants and cross-cumulants of the parent non-normal population. Despite this attention given to r, the magnitude of the bias and the variance of r are still relatively poorly understood for general bivariate nonnormal populations.…”
Section: Performance Of R 2 and R Under Bivariate Normalitymentioning
confidence: 99%
“…Although the correlation coefficient is commonly used to characterize the complicated rainfall spatial structure, accuracy and reliability associated with its estimation are not completely resolved (Stedinger 1981;Shimizu 1993;Lai et al 1999;Ciach and Krajewski 2006). The correlation using traditional Pearson formula can be overestimated in some cases for which rain rates follow non-normal distributions (Kowalski 1972).…”
Section: B Correlation Structurementioning
confidence: 99%