/Forecast-Methods-For-Google-Trends

Shiny application to test Forecast Methods in Google Trends

Primary LanguageHTML

Forecast-Methods-For-Google-Trends

This shiny application is for testing some simple forecasting models. We used the Google Trends database for this application using the "gtrendsR" package.

https://fabiorcampos.shinyapps.io/forecast_methods_for_google_trends_analysis/

Instructions

To use this application, enter a keyword that interests you, select the forecast template and click on "Run Code".

Results

  1. Plot Output: See the hits for the keyword searched in Google Trends and the forecasting for this search. The black curve within the predictive area is the training curve.
  2. Summary: Indicates the calculated errors for the applied forecast. This app will have new updates inserting new templates and views.

Packages

Philippe Massicotte and Dirk Eddelbuettel (2017). gtrendsR: Perform and Display Google Trends Queries. R package version 1.9.9.0.https://github.com/PMassicotte/gtrendsR

Hyndman RJ (2017). forecast: Forecasting functions for time series and linear models. R package version 8.1, URL:http://github.com/robjhyndman/forecast.

Adrian Trapletti and Kurt Hornik (2017). tseries: Time Series Analysis and Computational Finance. R package version 0.10-42.