_SLIDE_1_
Regression Analysis of International Tourism Revenue
_SLIDE_2_
Contents
Introduction
Why do we choose the topic
Source of data
Variables
Process
Verification of availability about the collected data
Model specification
Checking the assumptions of model
Detection influential point
Checking regression results
Variable Selection
Final Model
Conclusion
_SLIDE_3_
How can we attract foreign tourists consistently
Which factors affect the tourism receipts
How can we improve these factors
Introduction
Data sources
_SLIDE_4_
Variable
Type
Description
Y
Revenue
International tourism receipts
X1
Climate
Qualitative
Whether the nation has temperate climate or not
X2
Exchange rate
Quantitative
X3
Territory
Quantitative
X4
World match
Qualitative
Whether the nation have hosted international match
X5
Airports
Quantitative
The number of airports
X6
Tuberculosis
Quantitative
The number of tuberculosis patient
X7
¡¦(»ý·«)
|
d data
Process -
X2
lnX2
X3
lnX3
After Log transformation of X2&X3
_SLIDE_8_
1. Verification of availability about collected data
Process -
Checking correlation between predictors by coefficients of correlation
Between X3 and X5. and between X7 and X9, there are quite big coefficient of correlation (about 0.65).
Doubt the linear relationship.
Need to check the plot
_SLIDE_9_
1. Verification of availability about collected data
Process -
X3-X5
X7-X9
Checking correlation between predictors by the scatter plot
Through the plots, we can find that there is not serious linear relationship
_SLIDE_10_
Process - 2. Model Specification
,
_SLIDE_11_
3. Checking the assumptions of model
Process -
x1
x2
x3
x4
x5
x6
x7
x8
x9
x10
x11
x12
X5, X6 and X9 plots don¡¯t seem to follow linear line.
violating in the normality assumption
Q-Q Plot
_SLIDE_12_
3. Checking the assumptions of model
Process -
Checking standar