KuiperTestStat function

Plots cummulative density for Kuiper test and computes confidence interval for Kuiper test stat.

Kuiper test statistic is a non parametric test for distribution equality and is closely related to KS test. Formally, the Kuiper test statistic is : [REMOVE_ME]$D*=\max_i\{F(X_i)-\hat{F(x_i)}+\max_i\{\hat{F}(X_i)-F(X_i)\} [REMOVE_ME_2]$

KuiperTestStat(number.trials, sample.size, confidence.interval)

Arguments

• number.trials: Number of trials
• sample.size: Sizes of the trial samples
• confidence.interval: Confidence interval expressed as a fraction of 1

Returns

Confidence Interval for KS test stat

Description

Kuiper test statistic is a non parametric test for distribution equality and is closely related to KS test. Formally, the Kuiper test statistic is :

Examples

# Plots the cdf for Kuiper Test statistic and returns Kuiper confidence
   # interval for 100 trials with 1000 sample size and 0.95 confidence
   # interval.
   KuiperTestStat(100, 1000, 0.95)

Author(s)

Dinesh Acharya

References

Dowd, K. Measuring Market Risk, Wiley, 2007.