Terminology

---

Dataset

A composite database, block model, general database, grid or Vulcan screen object.

Graphs

The results of a statistical analysis shown in graph format.

Statistics

The results of a statistical analysis shown as text (result tables).

Variable Pairs

Data variables that are plotted against each other.

---

Graph types

Box Plot

This plot type produces a simple graphical summary of a set of data, allowing you to display a measure of central location (the median), two measures of dispersion (the range and inter- quartile range), the skewness (from the orientation of the median relative to the quartiles) and potential outliers.

Char Histogram

This plot type produces a histogram showing the number of character values that match a specified list of character patterns.

Cumulative

This plot type produces a graph that displays the probability that a random variable will be less than the independent variable.

General

This option does not produce a graph, but generates univariate statistics. These include the average, standard deviation, median, skewness, geometric mean, harmonic mean, mean of logs, variance of logs, kurtosis, Sichel T, median, quartiles and a percentage distribution listing.

You can select as many variables as you want to analyse. Each variable is analysed separately. For example, if you have a composite database with an AU field and a AG field and also a block model with an AU field and an AG field, when you select these 4 variables, you will get 4 statistics, one for each variable.

Histogram

This plot type generates a histogram of the data selected. Options exist for linear, log and normal Gaussian scales on both the X and Y axis.

For a histogram you define intervals. The program displays a bar for each interval. The height of the bar depends on how many samples are in its interval. You can let the program choose intervals or you can define your own intervals by controlling the width of the intervals, the number of intervals, or by directly defining each interval.

You can select any number of variables. Graphs are produced for every variable.

Joint

This option does not produce a graph, but generates statistics of correlation between any number of variable pairs. The variables must have the same number of elements. Joint statistics include correlation coefficient, rank correlation and linear regression slope and intercept.

Line Plot

This plot type produces a graph similar to the Scatter plot, but points are connected by lines. Each data pair is plotted in sequence, which means that the data needs to be logically organised in the data file, for example time series data.

Multiline

This plot type produces several line graphs of selected variable pairs. The data points are connected by lines.

Multivariate

This option does not produce a graph, but generates statistics in which the linear regression (best linear fit) coefficients (slope and intercept) are calculated between one dependent and several independent variables.

Normal Probability

This plot type produces a graph comparing a distribution with a log normal distribution. It is similar to Cumulative except that the X axis is LOG scale and the Y Axis is Gaussian (log-normal) scale. If the graph is a straight line, then the distribution is said to be log-normal. The probability axis is displayed in units of standard deviation and labelled in units from 0 to 1.

Pie

Similar to the Histogram except that the number of samples in an interval controls the size of a wedge in a circular graph.

PP Plot

This plot type produces a graph for each variable pair. The graph compares two distributions by plotting corresponding percentages on each axis. The variable pair does not need to have the same number of elements and may come from different datasets.

A point on the graph is plotted by taking a sample value which is common to both datasets and finding the percentile in each distribution. This pair of values is plotted as a point on the graph.

QQ Plot

This plot type produces a graph similar to PP Plot. A point on a QQ graph is plotted by taking a percentile and finding which value corresponds to that percentile in each distribution. This pair of values is plotted as a point on the graph.

Scatter

This plot type produces a graph showing a series of point markers for each selected variable pair. Plots are produced for every combination.

Ternary

This plot type produces a triangle graph of three variables. Each variable is represented as being a proportion of the sum of the triplet data. All three variables must have the same number of values. Points near one corner of the triangle indicate that one variable is predominant in the three variables. Points near the centre of the triangle indicate samples in which the three components are in roughly the same proportion.