Bucketing and min-max bounds on plots

[Fasjul]

Histogram bounds and step are 0-100 and 1, fixed. This should be dynamic as it makes few sense for anayltes that can go over 100 or which require a higher precision to be meaningful (ex pH in blood).

Additionally: The x-axis of histogram should either span the range
min(data) <--> max(data) or lowerBound(Confidence interval of lower RI bound) <--> upperBound(Confidence interval of upper RI bound), depending which is bigger.

[T-Blatter]

Number of bins and width

There are two common ways, that bin widths (and total number of bins) of a histogram are automatically calculated:

1. Sturge’s formula: ${\displaystyle k=\lceil \log _{2}n\rceil +1, }$ (used in EXCEL) see also
2. Freedman–Diaconis' formula: ${\displaystyle h=2{\frac {\operatorname {IQR} (x)}{\sqrt[{3}]{n}}},}$ (used in Minitab / SPSS) see also

$n$ = sample size, $k$ = number of bins, $h$ = bin width, $IQR$ = Interquantile range of data $x$

The 1. requires the knowledge of the overall sample size ($n$) and to calculate the bin width it requires the min / max of the data. The 2. only requires the knowledge of the quantiles ( $Q_1$ and $Q_3$ ) to calculate the bin width.

[f-marino]

@Fasjul how can we reproduce the issue? the pH analyte does not give the screenshotted plot with the data we have