forecast_q: Quarterly forecast evaluation of vintages

In johannes97s/ifwtrends: Evaluating Google Trends data for economic forecasts

Quarterly forecast evaluation of vintages

Description

forecast_q() uses quarterly vintages generated by roll() to predict the value of another time series for the next quarter.

Usage

forecast_q(r_list, data, fd = T)

Arguments

r_list	A Google Trends time series in form of a tibble (or tsibble) which is used to predict the next quarter in data.
data	A tibble with some data you want to predict for the next quarter that is somewhat correlated with the data in r_list.
fd	A logical value to indicate if you want to use the first differences in r_list (true) or normal values (false).

Value

Returns a list with the the predicted data and the used model results.

Methods

Google Data should be log-transformed currently it uses a RIDGE-Regression but other models such as OLS, LASSO or PCA are available.

Usage example

Check the service_imports.R file under /vignettes for an real example and to understand how to use this function and its twin, forecast_m().

johannes97s/ifwtrends documentation built on Oct. 9, 2022, 7:01 p.m.