Open AccessAnalysis of Vegetable Price Fluctuation Law and Causes based on Lasso Regression Model
Author(s) -
Weige Yu,
Chunjiang Zhao,
Huarui Wu,
Cheng Yuan Peng
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1284/1/012002
Subject(s) - multicollinearity , econometrics , lasso (programming language) , regression analysis , linear regression , regression , statistics , mathematics , price fluctuation , economics , computer science , financial economics , world wide web
In recent years, the sharp price fluctuation of vegetables has attracted wide attention. In order to explore the causes of price fluctuation, the price fluctuation law of cucumber from 2010 to 2017 was deeply explored based on the empirical analysis of the price fluctuation law of vegetables. Most of the existing literatures use linear regression models to analyze the influence degree of various factors, but there are problems of multicollinearity among multiple variables and small influence degree of some variables, which affect the fitting accuracy. Lasso regression model is introduced to model and solve cucumber price data, and the influencing factors of price with smaller correlation are eliminated to obtain the main factors and their correlation. Compared with the least square method, the multicollinearity determination condition value is only 19.66 and the fitting coefficient reaches 0.8448, which proves that lasso regression model is suitable for vegetable price cause analysis and has better performance than traditional methods. It also provides basis for further vegetable price prediction.