Analysis of Correlation Between China's Gross Domestic Product and Fresh Food E-Commerce Transactions Scale 2014-2022 by Using Simple Linear Regression Model

Kai XuTheory 2023 · DOI

This paper explores the quantitative relationship between China’s macroeconomic growth and the rapid expansion of its fresh food e-commerce sector. By applying a simple linear regression model to data from 2014 to 2022, the author demonstrates that GDP is a powerful predictor of the scale of online fresh food transactions, providing a mathematical foundation for market forecasting.

TL;DR

The study quantifies the strong positive correlation between China's GDP and the fresh food e-commerce market, finding that GDP growth explains over 96% of the variance in transaction scale. Using a simple linear regression model, the author provides a predictive framework for estimating future market expansion based on national economic performance.

What Problem Does This Paper Address?

While the growth of fresh food e-commerce in China is widely recognized, much of the existing literature focuses on qualitative aspects—such as supply chain logistics, cold chain technology, or platform business models. There is a lack of simple, accessible mathematical models that link this niche sector's growth to broader macroeconomic indicators.

The author seeks to bridge this gap by determining if China’s Gross Domestic Product (GDP) can serve as a reliable independent variable to forecast the "Fresh Food E-Commerce Transaction Scale." This is particularly relevant given the increasing reliance on cold chain logistics and the shifting consumption patterns of China’s growing population.

Methods

The author employs Simple Linear Regression (SLR) using IBM SPSS software. Rather than simply fitting a trend line, the paper emphasizes the importance of the Gauss-Markov assumptions to ensure the model's validity:

  • Linearity: Confirmed through scatter plot analysis of GDP vs. Transaction Scale.
  • Independence of Observations: Verified using the Durbin-Watson test, which yielded a value of 1.019 (indicating acceptable independence for this type of time-series data).
  • Normality: Assessed via histograms of standardized residuals and Normal P-P plots.
  • Homoscedasticity: Checked through a scatter plot of standardized residuals against predicted values to ensure constant variance.

The model follows the standard form: $Y = \beta_0 + \beta_1X + \epsilon$, where $X$ is GDP and $Y$ is the transaction scale.

Datasets

The study utilizes a time-series dataset spanning nine years (2014–2022).

  • Source: The ISV (Independent Software Vendors) e-commerce database.
  • Variables: Annual China GDP (measured in Trillion Yuan) and Annual Fresh Food E-Commerce Transaction Scale (measured in Trillion Yuan).
  • Data Trends: During this period, GDP grew from approximately 64.36 trillion Yuan to 121.02 trillion Yuan, while the fresh food e-commerce market expanded from 0.029 trillion Yuan to 0.56 trillion Yuan.

Key Insights and Ideas

  • Remarkable Explanatory Power: The model achieved an $R^2$ of 0.961 (Adjusted $R^2 = 0.956$). This suggests that 96.1% of the variation in fresh food e-commerce transactions can be explained by changes in GDP alone.
  • The "Chain Reaction": The author posits that GDP growth leads to increased consumption and a larger consumer base, which in turn drives demand for cold chain logistics, ultimately scaling the fresh food e-commerce market.
  • Statistical Rigor at the Entry Level: A notable aspect of this paper is the author’s (a high school student) adherence to formal statistical testing (Durbin-Watson, P-P plots) which is often overlooked in basic correlation studies.

Key Results

The regression analysis produced the following predictive equation: $$\text{Transaction Scale} = 0.009 \times \text{GDP} - 0.603$$

  • Coefficient Interpretation: For every 1 trillion Yuan increase in China's GDP, the fresh food e-commerce market is expected to grow by approximately 9 billion Yuan (0.009 trillion).
  • Significance: The ANOVA results showed an F-statistic of 173.792 with a p-value $< 0.001$, confirming that the relationship is statistically highly significant.
  • Correlation: The Pearson correlation coefficient ($R$) was 0.980, indicating an extremely strong positive linear relationship.

Limitations and Future Work

The author identifies several areas for improvement:

  • Small Sample Size: With only nine data points (2014-2022), the model is sensitive to outliers and may not capture long-term cyclical trends.
  • Omitted Variables: The model relies solely on GDP. Future research could benefit from Multiple Linear Regression to include variables such as technological advancement indices, climate impact, or smartphone penetration rates.
  • Data Scope: The author plans to collect pre-2014 data to improve the model's historical accuracy and predictive robustness.

Summary updated

Abstract

Along with the gradual increase of China's Gross Domestic Product (GDP) and the technological advances in the era of big data, the grain production has reached a new high and the sales scale of fresh food e-commerce transactions has expanded. This paper will use the parameter estimation of linear regression model and model test to analyze. The relationship between China's GDP and the sales scale of fresh food e-commerce transactions in recent years is analyzed by using simple linear regression model, and forecasts are made for the future. A linear equation for China's GDP as well as the sales scale of fresh food e-commerce transactions is fitted using parameter estimation. Homoscedasticity, independence of observations and normality were performed on this linear equation. The results show that the use of simple linear regression models can be used to analyze and predict the future trends of China's GDP and Fresh Food E-Commerce Transaction Scale.
BibTeX
@inproceedings{xu2023analysis,
  author = {Kai Xu},
  title = {{Analysis of Correlation Between China's Gross Domestic Product and Fresh Food E-Commerce Transactions Scale 2014-2022 by Using Simple Linear Regression Model}},
  booktitle = {Theory},
  year = {2023},
  doi = {10.47297/taposatwsp2633-456907.20230402},
}
Figure