general_plot: Create Scatterplot of Data (to visualize Correlations)

In MarcoNiemann/kaggle_house: Analysing House Prices in Ames, Iowa and Build a Sales Price Prediction Model

Description

The function creates a scatterplot of the complete dataset given by the variable and saves it in a PDF file. The lower panel makes use of panel.smooth to create the usual scatterplots for pairs of two variables but also adds some kind of a 'regression line' to the plot to increase the understandability. The upper panel uses a custom panel that shows of the numeric correlation values (highlighted with different font sizes and colors).

Usage

general_plot(x, name, threshold = 0.7, pdf = T, o = T,
  cor.met = "pearson")

Arguments

x	Dataset to be analysed.
name	Name that will be used in case a PDF will be produced.
threshold	The minimum correlation that two features must have to be considered further. If a feature does not have a correlation greater than the threshold with at least two other features it will be excluded from the correlation plot. Defaults to a value of 0.7.
pdf	Switch variable that decides whether the plot will be output to PDF or not. Defaults to PDF output.
o	Boolean switch to decide whether the method is allowed to print progress output to stdout. Defaults to true.
cor.met	Methodology used to compute the correlation. Can be "pearson", "spearman" oder "kendall". Defaults to "pearson".

MarcoNiemann/kaggle_house documentation built on May 7, 2019, 2:50 p.m.