Free radius script

Usage - Example:

• R = CRR(S) computes the radius of the mean-centered interval, circle, or sphere with 95% probability given S, which is either a vector of standard deviations or a covariance matrix from a multivariate normal distribution.
• R = CRR(S,P) computes the confidence region radius with probability P instead of the default, which is 0.95.
• R = CRR(S,P,TOL) uses a quadrature tolerance of TOL instead of the default, which is 1e-15. Larger values of TOL may result in fewer function evaluations and faster computation, but less accurate results. Use [] as a placeholder to obtain the default value of P.
• R = CRR(S,P,TOL,M) performs a bootstrap validation with M normally distributed random samples of size 1e6. Use [] as a placeholder to obtain the default value of TOL.
• R = CRR(S,P,TOL,[M N]) performs a bootstrap validation with M normally distributed random samples of size N.

Enhancements:

• Minor revisions
  

Requirements:

• MATLAB 6.5 (R13)

Major Features:

1. Easy usage: only a few lines of code are necessary to use a chart. No complex specification of data points, labels or grids necessary. This is done per default and may optionally be configured.
2. Multiple traces with different behavior may be added to a single chart.
3. Performance. JChart2D is designed for dynamic rendering. Rescaling is done per axis only and only if bounds are exceeded. New points within bounds are scaled without affecting old scaling. Minimum and maximum bounds of traces are cached and updated upon events. Invisible traces are not scaled and don't affect bounds.
4. AWT is the bottleneck: Non-visible calculations (scaling, rounding, event-handling, label calculation) do not impose high load even if a point is added with bounds changes every 2ms. But enlarging the window does. So scaling is bound to painting and the adapting painting frequency is limited to a configurable amount of milliseconds. For rendering traces poly-lines are drawn by default (instead of many calls to draw-line for connecting points). No comparisons to other libraries available.
5. Automatic choice of the unit prefix (multiplier) according to the scientific unit system (nano , milli, tera, ..).
6. Label format for scale labels via java.text. Number Format or java.text.Date Format.
7. Chooseable display of grids, labels, labeled scale-points, coordinate tooltips, ... .
8. Toolbox of UI controls for chart and traces via pop-up menus and menu bar. See the applet demo. Tip: right-click on the label of the trace to get controls for it.
9. Traces may be rendered via lines, discs, dots (is a disc with radius 1) or filled polygons. Additionally strokes may be specified (only useful for lines I think). An open interface for further graphical representations of traces exists.
10. Different axis implementations (settable from UI): linear, log e and log 10.
11. Multiple axes on top, bottom, left and right side. Traces may be assigned to an x and an y axis that has been assigned to the chart and will be related to them.
12. Different view-port implementations: Zooming (Zoomable Chart), enforcing visible bounds even if data does not fill them,... .
13. Exact visualization of labeled scale-points. No label of a scale-point is ever rounded to the next current chosen decimal, instead the scale-point is shifted to a position which represents an exact decimal.
14. Fully resizable. The amount of labeled scale-points may increase on choosing a bigger size.
15. Deadlock safe. Even this is hard to claim because testing depends on run-time behavior of the current system, i did not encounter a deadlock situation while letting several threads modify traces all contained in the same chart (only the art painting system may get pretty busy).
16. Trace implementations with different behavior: unordered, ring-buffered, ordered, ojective, ...
17. Highly configurable error bar API.
18. Support for zooming (Zoomable Chart)
19. Charts may be saved as images (jpeg, gif, jpg,... depending on the image writers of the java VM implementation), a snapshot UI control allows this.