Robust Interrupted Time Series Toolbox

Assessing the impact of complex policy interventions.

Get Started

Features for your analysis

State-of-art statistical analysis
The toolbox adopts the robust statistical methods published by (Cruz, Bender, and Ombao, 2019) based on a formal statistical model that can identify a change-point after a policy intervention and quantify its uncertainty.
Visualize results and metrics
The software provides a high level of interactivity in each graphical result to contrast the differences in pre- and post- intervention in the mean values and correlations. Each graphic can be zoomed, shifted and saved independently.
Run everywhere
We developed and optimized the software guarantying that it can run smoothly in the three major operating systems: Windows, Mac OSX, and GNU/Linux.
Accelerate results analysis
Focus on the analysis and not in the processing part: we have implemented all the steps in the model through a visual interface to reduce the complexity of the model.
Data privacy
We are also concerned about the use of sensible information. All processing is performed on your own computer and no information is sent to the network.
Open source
The toolbox is free and open source, double-licensed under the MIT and GPLv3 licenses. You can use, redistribute, and integrate it into your projects without any obstacle.
Our software is still under beta testing, and in the meantime, the GitHub repository is not open.

Downloads

RITS-Toolbox 3.0.1b (final version under development)


Download a sample dataset
GitHub repository

Getting started

1. Load the dataset

KRITS can process CSV tables. The file should have three columns: the institute name, the registering date, and the registered value.

2. Choose the date range

The toolbox will plot the time series in the dataset. Sometimes, we will need just a specific interval to analyze. Use the bars to choose the processing start and end dates.

3. Choose the potential change-point intervals

The statistical model requires an interval of theoretical (or potential) change-points. You can select the range using the bars. Remember that longer intervals will need more time to process.

Once you are sure about the intervals, press ”Model the dataset” to start the processing.

4. Data summary

In this table, you will see a brief summary of the processing results: the difference on the slope, levels and noise that appears in your data.

5. Relevant results

Extending the data summary, in this table you will see the difference between the estimated and your theoretical change-point proposal. Details about the slope, intercept and noise estimates are also presented along with the confidence intervals.

6. Explore the dataset: fitting and log-likelihood

Click on Menu View>Estimated mode to see the time series differences between the fitted time series with the change points. If you are wonder about the log-likelihood of the model, scroll down to see the plot.

Each graphic appears in an independent plot that can be zoomed and saved independently.

7. Explore the dataset: residuals and auto-correlation

Click on Menu View>Residuals to see the residuals of the fitted model (after and before the change-point) and their autocorrelation function according to the details expressed in the paper.

8. Read and save the report

Click on Menu Reports>Complete report. The toolbox collects all results data of the model through tables and plots in this large sheet. Click on DOCX report to save it in a MS Word 2017 format.

Note that due to the number of images, it could take a few seconds to load during the first time.

9. Get access to the documentation

The theoretical description of the model is also included in the toolbox. Click on Menu Help>Model description to read the paper and check its theoretical details.

Documentation

We are working on the documentation (and the paper) of the software. In the meantime we can refer to our previous papers to consult the statistical model.

Papers

For additional information, consult the following papers:

doi:10.1002/sim.8067
Cruz, Maricela & Gillen, Daniel & Bender, Miriam & Ombao, Hernando.
Statistics in medicine 38.10 (2019).
doi:10.1002/sim.7443
Cruz, Maricela & Bender, Miriam & Ombao, Hernando.
Statistics in medicine 36.29 (2017).

Research group

This project is handled by the Biostatistics Research Group at the King Abdullah University of Science and Technology

KAUST Biostatistics homepage

The software toolbox is developed by OPConsulting.

If you have any doubt about the model, or if you want us to know that the toolbox was useful to you, please contact Prof. Hernando Ombao.

If you have some issue in the installation, or if you discover a bug, please send us an email here: issues@op-consulting.tech or report an issue in our GitHub page.