Supervised Learning Algorithms for Sales Projection of Ecuadorian Shrimp with Python Programming Language
This research develops the best approximation for the non-linear projection of sales of a shrimp company listed on the Stock Exchange, in contrast to published corporate linear estimates. It starts from the search for data through a SWOT of the variable of interest: average price of Ecuadorian shrim...
| Autor: | |
|---|---|
| Tipo de recurso: | artículo |
| Estado: | Versión publicada |
| Fecha de publicación: | 2022 |
| País: | Ecuador |
| Institución: | Universidad Tecnológica Equinoccial |
| Repositorio: | Revistas Universidad Tecnológica Equinoccial |
| Idioma: | español |
| OAI Identifier: | oai:ojs.ute.edu.ec:article/996 |
| Acceso en línea: | https://revistas.ute.edu.ec/index.php/economia-y-negocios/article/view/996 |
| Access Level: | acceso abierto |
| Palabra clave: | Python programming language Econometrics Supervised learning Sales projection Shrimp industry Econometría Aprendizaje supervisado Lenguaje de programación python Proyección de ventas Industria camaronera |
| Sumario: | This research develops the best approximation for the non-linear projection of sales of a shrimp company listed on the Stock Exchange, in contrast to published corporate linear estimates. It starts from the search for data through a SWOT of the variable of interest: average price of Ecuadorian shrimp, identifying the variables: explanatory of shrimp prices in the United States, the observed change of the dollar against the yuan, Ecuadorian exports, US imports of Indian shrimp, barrel of WTI crude oil and the FPI™ salmon price index, as the most influential interpreted by the result of an adjusted coefficient of determination of 0.807. The instrumentation of the econometric model evaluates the statistical indicators of three predictive supervised learning linear regression algorithms in the Python programming language, with Ridge being the model with the lowest mean square error equal to 0.274. Based on five-year assumptions with Ridge, sales are forecast from 2021 to 2025, correlating the variables historical revenue of the shrimp company versus the average price of shrimp through polynomial interpolation, comparing both resulting trend lines showing that the expected revenues maintain a behavior non-linear according to its historical performance. |
|---|