Computes confidence intervals on regression on order statistics (ROS) mean

Uses ROS model output from the NADA package and computes the Zhou and Gao 1997 modified Cox’s method two-sided confidence interval around the mean for a lognormal distribution. Computes a t-interval for a gaussian ROS model output.

ROSci(cenros.out, conf = 0.95, printstat = TRUE)

Arguments

cenros.out

an ROS model output object (see details)

conf

Confidence coefficient of the interval (Default is 0.95)

printstat

Logical TRUE/FALSE option of whether to print the resulting statistics in the console window, or not. Default is TRUE.

Value

Prints a lower (LCL) and upper (UCL) confidence interval based on the conf provided (Default is 95%)

Details

This function uses an ROS model output based on the ros function in the NADA package. The lognormal distribution is the default for the NADA package but a gaussian distribution is optional here. For more detail on ROS modeling see the ros help file (?NADA::ros).

For implementation of ROSci(...) see the examples below.

References

Helsel, D.R., 2011. Statistics for censored environmental data using Minitab and R, 2nd ed. John Wiley & Sons, USA, N.J.

Lee, L., Helsel, D., 2005. Statistical analysis of water-quality data containing multiple detection limits: S-language software for regression on order statistics. Computers & Geosciences 31, 1241–1248. doi: 10.1016/j.cageo.2005.03.012

Zhou, X.-H., Gao, S., 1997. Confidence Intervals for the Log-Normal Mean. Statistics in Medicine 16, 783–790. doi: 10.1002/(SICI)1097-0258(19970415)16:7<783::AID-SIM488>3.0.CO;2-2

See also

NADA::ros

Examples

data(Brumbaugh)
myros <- NADA::ros(Brumbaugh$Hg,Brumbaugh$HgCen)

summary(myros)
#> 
#> Call:
#> lm(formula = obs.transformed ~ pp.nq, na.action = na.action)
#> 
#> Residuals:
#>      Min       1Q   Median       3Q      Max 
#> -0.23282 -0.05691 -0.01930  0.03697  0.59211 
#> 
#> Coefficients:
#>             Estimate Std. Error t value Pr(>|t|)    
#> (Intercept) -1.54719    0.01031 -150.11   <2e-16 ***
#> pp.nq        0.99735    0.01213   82.21   <2e-16 ***
#> ---
#> Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#> 
#> Residual standard error: 0.1088 on 116 degrees of freedom
#> Multiple R-squared:  0.9831,	Adjusted R-squared:  0.983 
#> F-statistic:  6759 on 1 and 116 DF,  p-value: < 2.2e-16
#> 

# ROS Mean
mean(myros$modeled)
#> [1] 0.3555983

# 95% CI around the ROS mean
ROSci(myros)
#> Assuming a lognormal distribution 
#>         LCL       UCL
#> 1 0.2883598 0.4385151