dataprep (version 0.1.5)

percplot: Plot the top and bottom percentiles of each selected variable

Description

The top and bottom percentiles of selected variables calculated by percdata can be plotted by percplot that offers a vivid check of possible outliers. It uses reshape2::melt or dataprep::melt to melt the data and uses ggplot2 and scales to plot the data.

Usage

percplot(data, start = NULL, end = NULL, group = NULL, ncol = NULL,
diff = 0.1, part = 'both')

Arguments

data

A data frame to calculate percentiles, from the column start to the column end.

start

The column number of the first variable to calculate percentiles for.

end

The column number of the last variable to calculate percentiles for.

group

The column number of the grouping variable. It can be selected according to whether the data needs to be processed in groups. If grouping is not required, leave it default (NULL); if grouping is required, set group as the column number (position) where the grouping variable is located. If there are more than one grouping variable, it can be turned into a longer group through combination and transformation in advance.

ncol

The total columns of the plot.

diff

The common difference between quantile's probs. Default is 0.1.

part

The option of plotting bottom and/or top percentiles (parts). Default is 'both', or 2 for both bottom and top parts. Setting it as 'bottom' or 0 for bottom part and 'top' or 1 for top part.

Value

Top (highest or greatest) and bottom (lowest or smallest) percentiles are plotted.

0th

Quantile with probs = 0

0.1th

Quantile with probs = 0.001

0.2th

Quantile with probs = 0.002

0.3th

Quantile with probs = 0.003

0.4th

Quantile with probs = 0.004

0.5th

Quantile with probs = 0.005

99.5th

Quantile with probs = 0.995

99.6th

Quantile with probs = 0.996

99.7th

Quantile with probs = 0.997

99.8th

Quantile with probs = 0.998

99.9th

Quantile with probs = 0.999

100th

Quantile with probs = 1

Details

Four scenes are considered according to the scales of x and y axes, namely the ranges of x and y values. For example, the code, sd(diff(log(as.numeric(as.character(names(data[, start:end])))))) / mean(diff(log(as.numeric(as.character(names(data[, start:end])))))) < 0.1 & max(data[, start:end], na.rm = T) / min(data[, start:end], na.rm = T) > = 10^3, means that the coefficient of variation of the lagged differences of log(x) is below 0.1 and meanwhile the maximum y is 1000 times greater than or equal to the minimum y.

References

1. Example data is from https://smear.avaa.csc.fi/download. It includes particle number concentrations in SMEAR I Varrio forest.

2. Wickham, H. 2007. Reshaping data with the reshape package. Journal of Statistical Software, 21(12):1-20.

3. Wickham, H. 2009. ggplot2: Elegant Graphics for Data Analysis. http://ggplot2.org: Springer-Verlag New York.

4. Wickham, H. 2016. ggplot2: elegant graphics for data analysis. Springer-Verlag New York.

5. Wickham, H. 2017. scales: Scale Functions for Visualization. 0.5.0 ed. https://github.com/hadley/scales.

6. Wickham, H. & Seidel, D. 2019. scales: Scale Functions for Visualization. R package version 1.1.0. https://CRAN.R-project.org/package=scales.

See Also

dataprep::percdata and dataprep::melt

Examples

# NOT RUN {
# Plot
percplot(data,5,65,4)

# Plot
percplot(data1,3,7,2)
# }