Bibliographic citations
Perez, C., Pimentel, R. (2023). Índices globales de búsquedas online acerca del zingiber officinale como predictores de las exportaciones peruanas de jengibre fresco (Partida arancelaria 091011) al resto del mundo 2013-2021: Un enfoque basado en redes neuronales [Tesis, Universidad Peruana de Ciencias Aplicadas (UPC)]. http://hdl.handle.net/10757/667914
Perez, C., Pimentel, R. Índices globales de búsquedas online acerca del zingiber officinale como predictores de las exportaciones peruanas de jengibre fresco (Partida arancelaria 091011) al resto del mundo 2013-2021: Un enfoque basado en redes neuronales [Tesis]. PE: Universidad Peruana de Ciencias Aplicadas (UPC); 2023. http://hdl.handle.net/10757/667914
@misc{renati/405241,
title = "Índices globales de búsquedas online acerca del zingiber officinale como predictores de las exportaciones peruanas de jengibre fresco (Partida arancelaria 091011) al resto del mundo 2013-2021: Un enfoque basado en redes neuronales",
author = "Pimentel Vignes, Raphael",
publisher = "Universidad Peruana de Ciencias Aplicadas (UPC)",
year = "2023"
}
This research aimed to determine whether global online search indices for ginger (Zingiber officinale) were predictors of Peruvian monthly exports of unground and unpulverised ginger to the rest of the world during the period 2013-2021, using a neural network approach. Indexes representing online searches for a specific term are considered a form of Big Data. It is possible to obtain them with Google Trends. Models that associate online searches with an eventual purchase can be associated with the Theory of Planned Behaviour. However, the background research showed that this was significant for durable or high-value goods, while for food the results were not very auspicious. Nevertheless, after analysing 108 monthly series, the authors of this research found that online searches for ginger were indeed predictors of Peruvian ginger exports. This was achieved by adding the different dimensions of online searches to a base model that included only one year's lags. The latter achieved an accuracy of 66.15%, while the version that added YouTube and image searches about ginger recorded an accuracy of 80.84%. Ordinary web searches did not contribute to the accuracy of the final model. In addition, when linear regression was used in a complementary way, direct and significant relationships between web and YouTube searches were evident, but the same was not true for the indices associated with image searches.
This item is licensed under a Creative Commons License