Empirical distribution

Counting and locating the values

Choose analysis.statistics.distribution. Set Numeric variable to voltage, Explicit bin rule to count, Bin count (count mode) to 6 and the prefix to guide.distribution.baseline. Untick Create group-mean graph for the numeric run. With the box ticked the provider requests the histogram and the ECDF.

The provider’s initial output prefix is statistics.distribution. The exercise uses a dedicated name.

Configuration in the GUI

Bin rules

count and Bin count (count mode)
Uses an integer number of equal intervals over the range of the finite data. The initial value is 10. The other two interval definitions are inactive for this calculation.
width and Bin width (width mode)
Uses a positive width in the units of X. The initial value is 1. Try 3 for voltage in V. The last interval covers the observed maximum.
edges and Bin edges, semicolon-separated (edges mode)
Write at least two edges in strictly increasing order, for example 8;12;16;20;25. The last edge belongs to the last bin. With edges that cover only part of the distribution read Outside bins. A decreasing list or one with duplicates prevents the processing.
Quantile probabilities, semicolon-separated
Enter probabilities in the interval [0, 1]. The initial value is 0;0.25;0.5;0.75;1. Try 0.1;0.5;0.9 to estimate the tenth percentile, the median and the ninetieth percentile. Use the decimal point.

Reading three complementary Datasets

.bins contains edges, centres, counts and frequencies. The relative and cumulative frequencies in any case use all valid observations as the denominator. .ecdf lists the 11 distinct support values of voltage in the example and the fraction of observations less than or equal to each value. .quantiles returns the estimated position for the requested probabilities, using type 7 interpolation and regardless of the bin rule.

In the report distribution_report.txt, 18 valid rows fall into the 6 bins and the initial probabilities give 8, 10, 12.5, 15.75 and 22 V. The ECDF has a dimensionless Y axis between 0 and 1 and remains defined even if some values fall outside the edges chosen for the histogram.

Result in the GUI

The report shows six bins, 18 valid observations and the five quantiles.
The report shows six bins, 18 valid observations and the five quantiles.

Download the report for this exercise.

Plot in Plots

Empirical distribution function of voltage in Plots. The ordinate reaches 1 at the observed maximum.
Empirical distribution function of voltage in Plots. The ordinate reaches 1 at the observed maximum.

For output IDs, selection and publishing criteria see Options shared by the providers. To look up or open the result go to Tables, reports and plots.