Frontiers in Computing and Intelligent Systems 
ISSN: 2832-6024 | Vol. 3, No. 1, 2023 

 

6 

Research on Wildlife Trade Based on SARIMA and 
Multiple Regression 

Zhihong Liao a, Jinyuan Song b and Jingyu Yang c 

Sichuan University, Chengdu 610207, China. 

a1732407275@qq.com, b1097537934@qq.com, c1733942830@qq.com 

 

Abstract: At the start, this paper firstly converges the data set given in the topic to count out the data from 1990-2021. After 

that, uses the count function of MATLAB on the pd library to statistically analyze the family data, so as to get the greatest number 

of wild animal groups and species transactions, and concludes that: among the species varieties, the category of Cercopithecidae 

has the most number, followed by Felidae, Cebidae. among the specific genera of animals, Macaca had the highest number, 

followed by Papio and Potos. Next, this paper analyzes the statistical options so as to obtain the most important trade purposes, 

and concludes that: the analysis of the global wildlife trade records from 2003 to 2021 shows that the purposes of wildlife trade 

are the most used for wildlife trade, zoo use, and circus performances. After that, this paper uses the number of trade imports and 

exports over the years as the basic data to measure the state of the trade market, and later analyzes it by constructing the growth 

rate of trade transactions as an indicator to assess the state of the wildlife trade market. The concludes that: with the change of 

time, the economy was relatively stable in the pre-wild trade market, and there was a market turmoil in 2012, and then it stabilized. 

Market turbulence was observed in both 2018-2022, with a depressed status quo in the wild trade market. Then, this paper will 

use multiple regression analysis to analyze the number of import and export transactions of wildlife trade each year and the 

degree of epidemic, so that the correlation between the two can be judged by the regression coefficient, and the conclusion is: 

the severity of the epidemic will indeed have a certain impact on the wild trade market, and this relationship shows a typical 

negative correlation, that is, the higher the severity of the epidemic, the lower the import and export The lower the severity of 

the epidemic, the higher the import and export trade in the wild trade market. Last, the annual wildlife trade import and export 

quantities are used as the basic data, and the total global GDP is collected as an indicator of the global economy, so that a multiple 

regression model is constructed to analyze the relationship between the global wildlife trade import and export quantities and 

the total global GDP, and concludes that: based on this analysis, this paper concludes that the global wildlife trade ban needs to 

be adhered to at least for a long period of time, and that the wildlife trade does affect the global economic situation.  

Keywords: S-ARIMA; Multiple regression; Wildlife Trade; MATLAB. 

 

1. Background 

In the long history of human development, there are traces 

of wild animals, which are closely related to the survival of 

human beings. In primitive society, wild animals were 

important sources of food and materials for human beings. 

This stage was called dependent utilization [1]. 

Gradually, humans learned to domesticate wild animals 

into poultry and domestic animals. This stage is called 

transformational utilization. In the modern society, people can 

hunt wild animals on a large scale and the demand for wild 

animals is very strong, so there has been a large-scale 

commercial trade in wild animals. In this stage, many species, 

such as the dodo and the North American traveling pigeon, 

were hunted to extinction due to large-scale commercial trade 

demand. People gradually realize that the uncontrolled and 

predatory use of wild animals is unsustainable, and relevant 

laws and regulations are needed to regulate the trade of wild 

animals. So we began to make a transition to the stage of 

coordinated utilization of wild animals. 

However, at this stage, a new study found that illegal 

wildlife trade is having a far-reaching negative impact on 

biodiversity. Researchers report that many species, including 

those living in the Reserve, are threatened by hunting and 

trapping to meet trade needs [2]. 

"This study adds to the evidence that wildlife trade is a 

major threat." Said Scott Robertson, an environmentalist with 

the Wildlife Conservation Society. Analysts estimate that the 

annual income of illegal wildlife trade is between US $5 

billion and US $20 billion, involving tens of millions of 

individuals who capture or kill thousands of species of 

animals, while about 150 million families rely on eating or 

selling wild animals for a living. 

David Edwards, a conservation biologist at the University 

of Sheffield in the United Kingdom, said that for decades, 

many environmentalists believed that wildlife trafficking was 

making certain species extinct, but some believed that "trade 

is usually sustainable". 

In order to achieve coordinated utilization, we have 

introduced some laws and regulations. First, the international 

wildlife trade management method – the Convention on 

International Trade in Endangered Wildlife, also known as the 

Washington Convention, CITES Convention. The core spirit 

of the Convention is to manage the international trade of wild 

animal and plant species by means of species classification 

and licensing, so as to achieve the purpose of protecting wild 

animals. 

2. Assumptions 

2.1. Assumptions 

1. It is assumed that in this paper, after data pre-processing, 

there is no anomalous data to interfere with the problem. 

2. Assume that in this paper, all the data are smooth, so the 

global wildlife trade import and export quantities at any 

moment can be obtained. 



 

7 

3. Assume that in this paper, the influence of other factors 

on the global wildlife trade is not considered. 

4. Assume that in this paper, the impact size of the epidemic 

level can be expressed by the basic regeneration number R0. 

5. It is assumed that in this paper, the time of onset of the 

epidemic is calculated from the very beginning of the onset. 

2.2. Notations 

Table 1. Symbol definition table 

Symbol Meaning 


 

which is the standard deviation. 

X
 

the standard data 


 
the mean of the dataset 

𝑦𝑡  Observations at period 𝑡 
𝛿 Constant term for time series 

forecasting 
𝜇𝑡 Random error 

𝑦𝑡−1 ⋯ 𝑦𝑡−𝑝 lagged time series 

𝜃1 ⋯ 𝜃𝑝 Coefficients of the regression model 

3. Question Analysis 

3.1. Model establishment and solution of 

problem 1 

For question one, the question asks for an analysis of the 

data set provided to find the most traded species and groups 

of wild animals. For this problem, this paper firstly combines 

the data set given in the title and calculates the data from 1990 

to 2021. After that, this paper screens the data set, selects the 

living wild animals at the present stage as the sample data, 

and uses MATLAB to carry out statistical analysis on the 

count function in pd library to obtain the maximum number 

of wild animal groups and species traded. Finally, this paper 

makes a theoretical analysis and verification of the trading 

animal groups and the maximum number of species trading, 

so as to prove that the conclusions obtained in this paper are 

consistent with the reality. 

Based on this, the flow chart of the first question is drawn 

in this paper as follows: 

 

Fig.1 The flow chart for solving Problem 1 

3.1.1. Statistical model of vulnerable animal population 

and species 

Based on the above analysis, this paper firstly pre-

processes the data set provided by the sponsor. Then, the data 

set from 1990 to 2021 was used to count the total   number 

of entries in each year, as shown in the figure below. 

 

Fig.2 Project maps for each year 

According to the above figure, the trend of the project chart 

in each year shows that the number of wild trade transactions 

decreases sharply with the increase of years. Further, the 

family data in the data are collected for statistical analysis, 

and the results are as follows: 

 

Fig.3 Ranking of species number transactions in wild trade markets 
from 1990 to 2021 

 

Fig.4 Ranking of animal group transactions in the wild trade 
market from 1990 to 2021 

According to the analysis in Figure 4.1 and 4.2, 

Cercopithecoidea has the largest number of species, followed 

by Felidae and Cebidae. In the specific genus of animals, 

Macaca is the most abundant, followed by Papio and Potos. 



 

8 

3.2. Model establishment and solution of 

problem 2 

In the second question, the purpose of trade of these 

animals is analyzed. For this problem, this paper considers 

that the Purpose option of the attached data is calculated on 

the basis of the screening of the first question, so as to draw 

the statistical distribution map. Further, this paper analyzes 

the statistical options, so as to get the main purpose of the 

transaction, and explore its reasons, so as to analyze the direct 

direction of the purpose. 

Based on this, the flow chart of the second question is 

drawn in this paper as follows: 

 

Fig.5 The flow chart for Question 2 

3.2.1. Analysis on the purpose of wild trade 

To solve this problem, on the basis of the first question, this 

paper makes statistics on the choice Purpose stated in the data, 

and the results are shown as follows:  

 

Fig.6 Statistical chart of wild trade market transactions from 1990 
to 2021 

Table .2 Wild trade market Transaction purpose explanation table 

Code Explanation 

B Breeding in captivity or 

artificial 

E Propagation Educational 

G Botanical garden 

H Hunting trophy 

L Law enforcement / 

judicial / forensic 

M Medical (including 

biomedical research) 

N Reintroduction or 

introduction into the wild 

P Personal 

Q Circus or travelling 

exhibition 

S Scientific 

T Commercial 

Z Zoo 

Based on this, this paper analyzes the 3D pie chart above, 

and it can be seen that the darker the color, the more quantity, 

and the brighter the color, the less quantity. Through the above 

analysis, it can be seen that T, Z, Q and choice have the most 

purposes. In order to understand the specific meaning of each 

indicator, this paper explains the transaction purpose in the 

above figure through the search options in the attached table, 

and the results are shown in the table 2. 

Through analysis of the above table, it can be seen that T 

stands for wildlife trade, Z for zoo use, and Q for circus 

performance. 

Therefore, among the purposes of wildlife trade in this 

question, those used for wildlife trade, zoo use and circus 

performance are the most. 

3.3. Model establishment and solution of 

problem 3 

To answer question three, the author asks for an analysis of 

the trade situation in the past 20 years. First of all, by 

determining the data set, this paper learns that the time node 

of the past 20 years is 2003-2022. After that, this paper 

integrated the data set from 2003 to 2022, took the number of 

trade imports and exports over the years as the basic data to 

measure the state of the trade market, and then constructed the 

growth rate of trade transactions as the state index to evaluate 

the wildlife trade market, so as to analyze it. MATLAB 

software is used to draw a line chart of the trade growth rate, 

so as to observe the changes of the global trade market in the 

past 20 years. 

Based on this, the flowchart of the solution of the third 

question is drawn in this paper as follows: 

 

Fig.7 The flow chart for question three 

3.3.1. Trade status assessment model 

To solve this problem, this paper firstly collates the 

attached data and selects the data from 2003-2022 as the data 

set. 

After that, this paper integrated the data from 2003 to 2022, 

and took the import and export in the attached data as the 

representation of the market state of wildlife trade. As we can 

see from the observation that there are multiple lines of data 

in each year's data, this paper uses the sum function for 

summation, so as to obtain the annual trade import and export 

quantity, as shown in the following table. 

Based on this, this paper draws the import and export 

quantity trend chart of wild trade market from 2003 to 2021, 

and the results are shown as follows: 

Based on this, this paper defines the growth rate of import 

and export volume, as shown below. 

Volume growth rate of import trade： 

𝑉_𝐼𝑖𝑛(𝑖 + 1) =
𝑉_𝐼𝑖𝑛(𝑖) − 𝑉_𝐼𝑖𝑛(𝑖 − 1)

𝑉_𝐼𝑖𝑛(𝑖 − 1)
 

Volume growth rate of export trade： 

𝑉_𝐼𝑜𝑢𝑡(𝑖 + 1) =
𝑉_𝐼𝑜𝑢𝑡(𝑖) − 𝑉_𝐼𝑜𝑢𝑡(𝑖 − 1)

𝑉_𝐼𝑜𝑢𝑡(𝑖 − 1)
 



 

9 

Based on this, this paper uses MATLAB software to 

calculate the growth rate of import and export volume, and 

the results are shown as follows: 

Table .3 Quantity of import and export of wild trade market from 
2003 to 2021 

Year 
Importer 

reported quantity 

Exporter reported 

quantity 

2003 8537 11885 

2004 11519 6987 

2005 8819 9600 

2006 5919 5849 

2007 3221 3216 

2008 3113 4088 

2009 3671 3507 

2010 5002 5233 

2011 2196 4200 

2012 2791 3665 

2013 2600 3575 

2014 3488 4054 

2015 2804 2483 

2016 2723 2916 

2017 4277 3793 

2018 6408 4174 

2019 1434 8018 

2020 598 6802 

2021 428 1333 

 

Fig.8 3D map of import and export quantity of wild trade market 
from 2003 to 2021 

 

Fig.9 Image of growth rate of import and export of wild trade 
market from 2003 to 2021 

According to the analysis in the figure above, with the 

change of time, the economy was relatively stable in the early 

stage of the wild trade market, and there was a market 

turbulence in 2012, and then it stabilized. In 2018-2022, there 

was market turmoil and a downturn in the wild trade market. 

3.4. Model establishment and solution of 

problem 4 

In response to question 4, we need to explore whether 

wildlife trade is related to the epidemic situation of major 

infectious diseases. For this problem, this paper first analyzes 

the annual import and export volume of wildlife trade. Then, 

the transmission time and epidemic degree of major infectious 

diseases are obtained by consulting relevant literature. Then 

this paper will use multiple regression analysis to conduct 

regression analysis on the annual import and export volume 

of wildlife trade and the epidemic degree, so as to determine 

the correlation between the two through regression coefficient 

Based on this, the flow chart for solving the fourth question 

in this paper: 

 

Fig.10 The flow chart problem 4 solution 

3.4.1. Analysis of Wildlife Market Trade Market and 

Epidemic Degree 

Based on this, this paper makes statistics on infectious 

disease events that have occurred in the historical process, and 

conducts multiple regression analysis on the import and 

export quantity of wild animal trade market. 

First of all, this paper analyzes the infectious disease events 

in history and the corresponding time, as shown in the 

following table. 

Table .4 Sequence of influenza virus events in 2003-2021 

Year Event 

2003 SARS 

2009 SWINE FLU 

2014 Ebola Virus 

2019 COVID-19 

Based on this, this paper analyzes the events of influenza 

virus and the corresponding import and export quota of wild 

market trade, and draws the image using matlab as follow. 

 

Fig.11 Influenza related to wild trade market from 2003 to 2021 



 

10 

According to the analysis of the above figure, when an 

influenza event occurs, the value of the global wild trade 

market will decline significantly, which will lead to a 

depression in the wild market trade. 

Based on this, this paper further introduces the basic 

regeneration number R0 to evaluate the extent of infectious 

diseases. R0 is an estimate of the transmission speed of a 

specific infectious disease in a specific population. It is 

usually calculated based on three parameters: 1) the duration 

of contact infection after infection, 2) the possibility of 

infection between infected individuals and susceptible 

individuals, and 3) the contact rate. 

Specifically, R0 refers to the average number of people that 

can be infected by one person. If R0＜1, it means that the 

disease is under control or not spreading fast; If R0 is 1, one 

person can spread to another on average; If R0>1, the disease 

can spread from one person (index) to a wider population. 

Therefore, this paper queries R0 of infectious diseases, as 

shown in the following table. 

Table .5 Sequence of influenza virus events in 2003-2021 

Year Event Basic 

regeneration 

number 

2003 SARS 1.08 

2009 SWINE FLU 1.6 

2014 Ebola Virus 1.9 

2019 COVID-19 5.7 

Based on this, the basic regeneration number in the 

remaining years without influenza is set as 0, so as to conduct 

regression analysis. Therefore, this paper introduces multiple 

regression analysis as follows. 

The multiple linear regression model predicts the linear 

relationship between the independent variable 𝑦  and 

multiple dependent variables  𝑥  and errors 휀 . Let 𝑦  be the 

dependent variable and the independent variable be 

𝑥1, 𝑥2, . . . , 𝑥𝑛  , then the multiple linear regression model is 

expressed as follows [8]: 

y = 𝛽0 + 𝛽1𝑥1+. . . +𝛽𝑛𝑥𝑛 + 휀 

Among them, β0, β1. . . , βn are unknown parameters, 

assuming random error ε： 

( ) 2,0)(  == VarE
 

The theoretical regression equation is 

𝐸(𝑦) = 𝛽0 + 𝛽1𝑥1+. . . +𝛽𝑛𝑥𝑛 

For practical problems, assuming that n  sets of sample 

data are obtained, the multiple regression equation can be 

expressed as: 

{
𝑌 = 𝑋𝛽 + 휀

𝐸(휀) = 0n, 𝑉𝑎𝑟(휀) = 𝜎2𝐼𝑛
 

while 

𝑌 = [

𝑦1

𝑦2

. . .
𝑦𝑛

] , 𝑋 = [

1, 𝑥11, 𝑥12, . . . , 𝑥1𝑛

1, 𝑥21, 𝑥22, . . . , 𝑥2𝑛

. . .
1, 𝑥𝑚1, 𝑥𝑚2, . . . , 𝑥𝑚𝑛

] , 𝛽 = [

𝛽1

𝛽2

. . .
𝛽𝑛

] , 휀 = [

휀1

휀2

. . .
휀𝑛

] 

Get an estimate of the vector β by least squares: 

𝛽
^

= (𝑋′𝑋)−1𝑋′𝑦 

The estimated model is: 

𝑦
^

= 𝛽
^

0 + 𝛽
^

1𝑥1+. . . +𝛽
^

𝑛𝑥𝑛 

The regression line fitting effect is determined by the 

proportion of the regression square sum SSR to the total 

square sum SST, that is, the coefficient of determination 

R2 =
SSR

SST
 . The closer it is to 1, the better the regression line 

fits the original data distribution. 

The estimated standard deviation represents the size of the 

error between the actual and predicted values. The formula is: 

𝑠𝑒 = √∑(𝑦𝑖 − 𝑦𝑖)
2

^

𝑛 − 𝑝 − 1
= √

𝑆𝑆𝐸

𝑛 − 𝑝 − 1
= √𝑀𝑆𝐸 

The smaller the value se , the smaller the difference 

between the actual value and the predicted value, and the 

closer the forecast is to the actual value. If se=0, it means that 

the actual value is equal to the predicted value. 

The significance test includes linear relationship test and 

regression coefficient test. 

The linearity test is as follows: 

If the null hypothesis ( H0: βI = 0 ) is established, the 

distribution of MSR/MSE obeys the F(p,n-p-1) distribution, 

is [9]: 

)1,(~
)1(

/
−−=

−−
= pnpF

MSE

MSR

pnSSE

pSSR
F

 
If the null hypothesis is rejected, it is determined that there 

is a significant linear correlation between the independent 

variable and the dependent variable. 

The regression coefficient test is as follows: 

𝐻0: 𝛽𝐼 = 0 

Tests statistic: 

𝑡 =
𝛽𝑖

^

𝑠
𝛽𝑖

^
=

𝛽𝑖

^

𝑠𝑒

√∑𝑥𝑖
2 −

1
𝑛

(∑𝑥𝑖)
2

~𝑡(𝑛 − 𝑝 − 1) 

If the null hypothesis is rejected, it means that there is a 

significant linear correlation between the independent 

variable and the dependent variable. 

Thus, the model is constructed by using multiple regression, 

and the basic regeneration number R0 is taken as the 

dependent variable, and the time, import and export trade 

volume of wild market are taken as the dependent variables. 

The results are as follows. 

R0 = −144.3764 + 0.0566 ∗ year − 1.92 ∗ 10−4 ∗ In −
3.74 ∗ 10−4 ∗Out 

Therefore, according to the analysis of the above 

relationship, if the coefficient in front of the relationship is 

larger, it means that the relationship between them is closer, 

otherwise, it means that the relationship between them is less 

obvious. For the coefficient of each index, if the coefficient 

of this index is positive, it means that there is a positive 

correlation between the independent variable and the basic 

regeneration number R0, and vice versa. 

Based on the above analysis, we can see that the severity of 

the epidemic does have some impact on the wild trade market, 

and this relationship shows a typical negative correlation, that 

is, the higher the severity of the epidemic, the lower the 

import and export trade volume of the wild trade market, and 

the lower the severity of the epidemic, the higher the import 

and export trade volume of the wild trade market. 

3.5. Model establishment and solution of 

problem 5 

As for problem 5, we need to judge whether wildlife trade 

is prohibited for a long time, and whether wildlife trade will 

have a huge impact on the economy and society, and explain 

the reasons. To solve this problem, this paper first uses the 



 

11 

annual import and export volume of wildlife trade as the basic 

data, and collects the total global GDP as the indicator to 

measure the global economy, so as to build a multiple 

regression model to analyze the data relationship between the 

import and export volume of global wildlife trade and the total 

global GDP, and according to the relationship between the 

three, Answer whether wildlife trade should be banned for a 

long time and its impact on the global economy. 

Based on this, the flow chart for solving Problem 5 in this 

paper: 

 

Fig.12 The flow chart of problem 5 solution 

3.5.1. Regression analysis of the global economy and 

wildlife trade 

Based on this, the relevant website 

(https://www.shujujidi.com/caijing/97.html)is inquired to 

obtain the data of global GDP over the years (in trillion 

dollars), and plotted the trend chart, as shown in the figure 

below. 

 

Fig.13 Global GDP trend from 1960 to 2021 

Based on the above figure, this paper uses the S-ARIMA 

time series prediction model to forecast the future trend of 

global GDP and the import and export volume of wild trade. 

Therefore, this paper first introduces the S-ARIMA time 

series prediction model, as shown below. 

With the widespread application of linear time series 

models, the models have been continuously improved. The 

SARIMA model is gradually developed on the basis of the 

autoregressive model (AR) and the moving average model 

(MA) [10]. In this problem, the concentration of CO2 can be 

regarded as a data series that changes over time, referred to as 

a time series. Therefore, the above model can be used to 

predict the future concentration trend of the time series, and 

the specific steps are as follows 

(1) Autoregressive model AR 

According to the relationship between current data and 

historical data, the autoregressive model can use historical 

data to predict future development trends. The autoregressive 

model has the following formula 

𝑦𝑡 = 𝛿 + 𝜃1𝑦𝑡−1 + 𝜃2𝑦𝑡−2 + ⋯ + 𝜃𝑝𝑦𝑡−𝑝 + 𝜇𝑡 

In the formula, yt is the observed value of period t, δ is 

the constant item, μt is the random error, yt−1 ⋯ yt−p is the 

lagged time series, θ1 ⋯ θp  is the regression model 

coefficient. 

(2) Moving Average Model MA 

If it is found that there is no correlation between the 

historical time series and the current time series, it is 

considered that the change is caused by external factors. At 

this time, the MA moving average model is used for 

prediction. The specific formula is as follows 

𝑦𝑡 = 𝜇𝑡+𝜇1 + 𝜃1𝜇𝑡−1 + 𝜃2𝜇𝑡−2 + ⋯ + 𝜃𝑝𝜇𝑡−𝑝 

(3) Autoregressive Moving Average Model ARMA 

If it is found that the current time series is affected by 

historical time series and external factors at the same time, the 

autoregressive moving average ARMA model is used for 

prediction. The specific formula is as follows 

𝑦𝑡 = 𝜃1𝑦𝑡−1 + 𝜃2𝑦𝑡−2 + ⋯ + 𝜃𝑝𝑦𝑡−𝑝 + 𝜇𝑡+𝜃1𝜇𝑡−1

+ 𝜃2𝜇𝑡−2 + ⋯ + 𝜃𝑝𝜇𝑡−𝑝 

(4) Autoregressive Difference Moving Average Model 

ARIMA 

Depending on how stationary the data is, predictive models 

can be constructed in different ways. Among them, the 

ARMA model is modeled for the stationary data, and the 

ARIMA model is calculated after the difference of the non-

stationary data. The mathematical expression of the 

difference is as follows, where ∆dyt = (1 − L)dyt  is the 

introduced d-order difference operator. 

wt = ∆dyt = (1 − L)dyt 

ARMA modeling is carried out after difference, and the 

formula is as follows, where wt  is the time series in a 

stationary state obtained after difference processing. 

yt = θ1yt−1 + θ2yt−2 + ⋯ + θpyt−p + μt+θ1μt−1 + θ2μt−2

+ ⋯ + θpμt−p 

(5) Autoregressive Moving Average Model SARMA 

If there is a seasonal change trend in the time series, the 

SARIMA model can be used for prediction. The model is 

simplified to SARIMA (p,d,q)(P,D,Q), and the specific 

formula of the model is as follows 

𝜃𝑝(𝐿)𝛷(𝐿𝑠)(1 − 𝐿)𝑑(1 − 𝐿𝑠)𝐷𝑋𝑡 = 𝜃𝑞𝛩(𝐿𝑠)𝜇𝑡 

(6) Testing of Time Series Forecasting Models 

Based on this, the timing test and stability test are carried 

out on the above-mentioned model, so that the accuracy of the 

model can be judged. 

Therefore, this paper forecasts the global GDP and the 

import and export volume of wild trade, as shown below: 

Table .6 2023-2025 S-ARIMA Time Series Forecasting Results 

Year Global 

GDP 

Import 

volume of 

wild trade 

Export 

volume of 

wild trade 

2023 96.7 334 1043 

2024 97.5 223 869 

2025 98.3 132 634 

Then, take the global GDP as the dependent variable Y, and 

take the import and export value and time of wild trade as the 

independent variable X to conduct multiple regression 

analysis. The results are as follows: 

Gobal GDP = −4620 + 2.3344 ∗ time − 6.31 ∗ 10−4 ∗ In
− 7.96 ∗ 10−4 ∗ Out 

According to the analysis of the above formula, the global 

GDP shows a typical proportional promotion relationship 

with time, while there is a typical negative correlation with 

global wildlife trade, that is, wildlife trade has an inhibitory 

effect on the growth of global GDP. 



 

12 

Therefore, through the analysis of the above equation, we 

can see that wildlife trade will have a negative impact on the 

world economy. However, through reading the relevant 

literature, we can see that a large part of the population in the 

global economy still depends on wildlife trade for survival. If 

we directly ban wildlife trade across the board, it may lead to 

economic bankruptcy and homelessness for many people. 

However, a lot of research and analysis show that under the 

current conditions, the emerging zoonosis may continue to 

grow at a constant rate, saying that wildlife trade is the main 

cause of the disease, and that those who rely on wildlife trade 

to survive have always suffered from infectious diseases, and 

the medical expenses have exceeded their own profit costs. 

Based on the above analysis, this paper believes that it is 

necessary to adhere to the global wildlife trade ban at least for 

a long time, and wildlife trade will indeed affect the global 

economic situation. 

4. Summary 

This article examines the relevance of wildlife trade, 

infectious diseases, and the global economy now that global 

epidemics are rampant. At the start, this paper firstly 

converges the data set to count out the data from 1990-2021. 

After that, uses the count function of matlab on the pd library 

to statistically analyze the family data, so as to get the greatest 

number of wild animal groups and species transactions. Next, 

this paper analyzes the statistical options, and concludes that 

the purposes of wildlife trade are the most used for wildlife 

trade, zoo use, and circus performances. After that, this paper 

uses the number of trade imports and exports over the years 

as the basic data to measure the state of the trade market, and 

later analyzes it by constructing the growth rate of trade 

transactions as an indicator to assess the state of the wildlife 

trade market. Then, this paper will use multiple regression 

analysis to analyze the number of import and export 

transactions of wildlife trade each year and the degree of 

epidemic, so that the correlation between the two can be 

judged by the regression coefficient. Last, the annual wildlife 

trade import and export quantities are used as the basic data, 

and the total global GDP is collected as an indicator of the 

global economy, so that a multiple regression model is 

constructed to analyze the relationship between the global 

wildlife trade import and export quantities and the total global 

GDP. 

References 

[1] Wang Hucheng, Du Wei, Wang Fei, Ge Bingjue. On the 
Challenges and Countermeasures of Administrative Law 
Enforcement of Wildlife Protection under the New Judicial 
Interpretation [J]. Jiangsu Forestry Science and Technology, 
2022,49 (04): 53-56. 

[2] Gui Yasheng. The judicial interpretation of wildlife protection 
in China needs to be revised [J]. Journal of Wildlife, 2019, 40 
(01): 217-220. DOI: 10.19711/j.cnki.issn2310-1490. 2019. 01. 
037. 

[3] Qin Tian, Ruan Xiangdong, Duan Zhaojun, Cao Jianping, 
Liang Junrong, Yang Jing, Jiang Yan, Shi Mang, Shen Yujuan, 
Xu Jianguo. Carry out research on wildlife microorganisms to 
deal with new infectious diseases in the future [J]. Disease 
Monitoring, 2021,36 (03): 209-213. 

[4] Zhou Ke, Hou Jiaru Improve the source of the rule of law for 
wildlife protection to prevent and control the spread of 
infectious diseases [J]. Law and Society, 2020 (03): 70-71. 

[5] Wu Wei Research Progress and Future Key Research Fields of 
Infectious Diseases of Wild Economic Animals [C]//. Special 
Papers of the Agricultural Branch of the 2005 National 
Academic Annual Meeting, 2005: 362-364. 

[6] Experts call for revising the wildlife protection law to 
completely prohibit wildlife trade [J]. Animal Husbandry 
Environment, 2020 (04): 1. 

[7] Zhao Dan. Animal and Environmental Organizations Call on 
the White House to Support the G20's Prohibition of Wildlife 
Trade [J]. China Forestry Industry, 2021 (Z1): 156. 

[8] Chen Ling, Li Zhenlong, Liu Tongxu, Su Rui, Sun Wei, Sun 
Wei. Multivariate regression analysis of weight and body size 
of Hu sheep at different ages [J/OL]. Chinese Journal of 
Animal Husbandry: 1-12 [2012-12-05]. DOI: 10.19556/j.0258-
7033.20220417-02. 

[9] Nong Yingxiong, Chen Zhibin, Huang Cong, Pan Jian, Liang 
Dong, Wei Yi, Li Zhe, Zhou Zhaofeng, Lu Ying. Factor 
analysis of tobacco growth based on multiple regression [J]. 
Hubei Agricultural Science, 2022,61 (19): 92-96+140.DOI: 
10.14088/j.cnki.issn0439-8114.2022.19.018. 

[10] Chen Kui, Dong Chenxue, Lu Jiayue, Ge Guoshu. Prediction 
of the number of AIDS cases in China based on ARIMA time 
series model [J]. China Primary Health Care, 2022,36 (03): 91-
93.