deltaforecast

An R package offering quick and easy prototyping for non-causal impact analysis.

https://github.com/shahabhishek1729/deltaforecast

Science Score: 44.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • CITATION.cff file
    Found CITATION.cff file
  • codemeta.json file
    Found codemeta.json file
  • .zenodo.json file
    Found .zenodo.json file
  • DOI references
  • Academic publication links
  • Academic email domains
  • Institutional organization owner
  • JOSS paper metadata
  • Scientific vocabulary similarity
    Low similarity (9.5%) to scientific vocabulary

Keywords

causalimpact impact-analysis r time-series time-series-analysis time-series-forecast time-series-forecasting
Last synced: 6 months ago · JSON representation ·

Repository

An R package offering quick and easy prototyping for non-causal impact analysis.

Basic Info
  • Host: GitHub
  • Owner: shahabhishek1729
  • License: other
  • Language: R
  • Default Branch: master
  • Homepage:
  • Size: 55.7 KB
Statistics
  • Stars: 1
  • Watchers: 1
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Topics
causalimpact impact-analysis r time-series time-series-analysis time-series-forecast time-series-forecasting
Created about 3 years ago · Last pushed over 2 years ago
Metadata Files
Readme License Citation

README.md

DeltaForecast

A package, written in R, offering quick and easy prototyping tools for non-causal impact analysis.

All you need to get started is a time-series and an event for which you want to measure the impact on the time series. For instance, perhaps you wish to measure the impact of a new school curriculum on students' test scores. Given a time-series dataset of the students' test scores in a school and the date the new curriculum was implemented, DeltaForecast can determine whether test scores improved (with statistical significance) or nor following the new curriculum.

NOTE: This package does not aim to identify causation - it serves as a preliminary assessment of whether or not any difference was perceived following an event, not necessarily as a result of that event. If you are interested in causation, CausalImpact provides this functionality if you have control time series to compare against. DeltaForecast builds on CausalImpact to use time-series forecasting in place of a synthetic control - i.e., models predict "what would have happened" in the absence of your event of interest, and compare this counterfactual against the observed/true data.

Getting started

```r library(DeltaForecast)

If you have a path to your data, you can pass this in to DeltaForecast with the path parameter

The following call tells DeltaForecast that the data is stored in "path/to/data.csv" (relative to the current

working directory, each individual row in the data represents a day of data (i.e., time in this dataset is measured

in days), the event of interest occurred 120 days (the temporal unit) after the beginning of the dataset, the

column storing the temporal data is called "Date" in the data, and the column storing the target value to be measured

is called "Scores".

impact <- DeltaForecast( path="path/to/data.csv", # Data is stored in "path/to/data.csv" event=120, # Event of interest occurred 120 days (frequency) from the start of the dataset frequency="D", # Each row in the dataset represents a single day temporalcol="Date", # The column storing the temporal data in our dataset is called "Date" targetcol="Scores" # The colun storing the target data in our dataset is called "Scores" ) ```

Owner

  • Login: shahabhishek1729
  • Kind: user

Citation (CITATION.cff)

cff-version: 1.2.0
message: "If you use this software, please cite it as below."
authors:
- family-names: "Shah"
  given-names: "Abhishek"
title: "DeltaForecast"
version: 1.0.0
date-released: 2023-01-17
url: "https://github.com/shahabhishek1729/DeltaForecast"

GitHub Events

Total
Last Year