otvPlots: Over time variable plots for predictive modeling (otvPlots)

In otvPlots: Over Time Variable Plots

Description

The otvPlots package uses data.table and ggplot2 packages to efficiently plot time series aggregated from large datasets. Plots of numerical variables are optionally returned ordered by correlation with date – a natural starting point for anomaly detection. Plots are automatically labeled if a variable dictionary is provided.

Details

Output files include:

• A PDF file of plots saved as outFl.pdf, with each individual page on one variable. Variables are plotted in the order indicated in the argument sortVars or sortFn. For each numerical variable, the output plots include

  • side-by-side boxplots grouped by dateGpBp (left),

  • a trace plot of p1, p50, and p99 percentiles, grouped by dateGp (top right),

  • a trace plot of mean and +-1 SD control limits, grouped by dateGp(middle right), and

  • a trace plot of missing and zero rates, grouped by dateGp (bottom right).

  For each categorical variable (including a numerical variable with no more than 2 unique levels not including NA), the output plots include

  • a frequency bar plot (left), and

  • a grid of trace plots on categories' proportions over time (right). If the variable contains more than kCategories number of categories, trace plots of only the largest kCategories will be plotted. If the variable contains only two categories, then only the trace plot of the less prevalent category will be plotted.

• CSV file(s) on summary statistics of variables, both globally and over time aggregated by dateGp. The order of variables in the CSV files is the same as in the PDF file.

  • For numerical variables, number of observations (counts), p1, p25, p50, p75, and p99 qunatiles, mean, SD, missing and zerorates are saved as outFl_numerical_summary.csv.

  • For categorical varaibles, number of observations (counts) and categories' proportions are saved as outFl_categorical_summary.csv. Each row is a category of a categorical (or binary) variable. The row whose category == 'NA' corresponds to missing. Categories among the same variable are ordered by global prevalence in a descending order.

License

Copyright 2017 Capital One Services, LLC Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0 Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.

See Also

Main function: vlm.

Selected supporting functions: PrepData, PrepLabels, OrderByR2.

otvPlots documentation built on May 1, 2019, 6:49 p.m.