tiểu luận kinh tế lượng FACTORS AFFECTING THE PRICE OF THE HOUSE

AbstractBackground: This paper looks at quantifiable and easily obtainable factors that affect housing price such as Age of the house, square of the house, the distance fromthe house to

FOREIGN TRADE UNIVERSITY INTERNATIONAL ECONOMICS FACULTY

I Abstract 1

II Introduction: 2

2.1 Basic Concept: 2

2.2 Why we carry out this survey and our expectation of the overall results? 2

III Literature review 3

3.1.Area 3

3.2 Location 3

3.3.Security 4

3.4 Social infrastructure 4

IV DESCRIPTION OF DATA 5

4.1.How did we collect the data? 5

4.2.Explaining variables 7

4.3 Using GRETL to analyze the variables 8

V ECONOMETRICS MODEL 9

5.1.Population regression function (PRF) 9

5.2.Sample regression function (SRF) 9

5.3.Expectation about the variables 9

VI RESULT ESTIMATION 11

6.1Run the Model with GRETL: 11

6.2.Interpretation of the GRETL model 11

6.3Rebustness Check 13

6.4Hypothesis testing of the coefficient number and variations 18

VII RESULT ANALYSIS AND POLICY IMPLICATION ……….22

VIII CONCLUSION………22

IX REFERENCES 23

I Abstract

Background: This paper looks at quantifiable and easily obtainable factors that

affect housing price such as Age of the house, square of the house, the distance fromthe house to university, whether the house has pool or not and whether the house hasfireplace or not , using data from our survey The goal of this analysis is to test therelationship between these five variables: Age, Sqft, Utown, Pool, Fplace using aregression model

Methods: A survey was conducted with people mainly in Ha Noi and Ho Chi Minh

City between September 17 and September 30/2019 The participants provided dataabout potential risk factors

Results: Among 1100 eligible participants, 1000 (90.9%) had refraction and

questionnaire data available All of the participants were surveyed with housingprice From that, we analyzed the information and then the result comes out:Distance to the university is the most important factor that affect housing prices

Conclusions: It can be concluded that age, acreage, location, utilities do affect, or at

least statistically so, the housing prices From that, the relevant people can rely on tomake the right decision in the future

II Introduction:

2.1 Basic Concept:

2.1.1 What is econometrics?

Literally interpreted, econometrics means “economic measurement”.

Econometrics is the application of statistical and mathematical theories in economicsfor the purpose of testing hypotheses and forecasting future trends It takes economicmodels, tests them through statistical trials and then compares and contrasts theresults against real-life examples

2.2.2 Why we have to study econometrics?

Econometrics is a set of research tools also employed in the business disciplines

of accounting, finance, marketing and management In today’s world, where there is

an insurmountable amount of data, econometrics is a vital skill to have It can help

us get a better understanding of the available data - which can help in mosteconomic decision-making processes That’s why it is used by social scientists,specifically researchers in history, political science and sociology, etc Econometrics plays an important role in such diverse fields as forestry, and inagricultural economics, Most economists use economic data to estimate economicrelationships, test economic hypotheses, and predict economic outcomes

“Econometrics fills a gap between being “a student of economics” and being “apracticing economist.” “

2.2 Why we carry out this survey and our expectation of the overall results?

First and foremost, the fact that we carry out this subject help us tremendouslywith our future careers when we have to apply what we learned into practice Forany student deciding to pursue the field of finance or financial consultant, againeconometrics plays a very important role To do this, econometrics can become quitehandy Hence, econometrics comes into use in some way or the other in almost allfields that a student will pursue in our career

The reason why we choose the topic: “Factors affecting the price of the house”

is quite simple The more Vietnamese economy develops , the higher demand for

house in center city is The number of households in the country is set to grow by3% by 2013, according to the Vietnam Housing Forecast 2013 from market researchcompany RNCOS The government has directed the Ministry of Construction tobuild more homes and US$173 million is being invested in 37 low cost housingprojects covering a total of 750,000 square meters A high amount of investment inthe Vietnam housing sector has resulted in soaring growth over the past few yearsespecially in cities like Hanoi and Ho Chi Minh City That is the generaldevelopment of developing countries and Viet Nam can not be outside of that trend.

So, that is the reasons motivated us to go into this field and topic as well

III Literature review

Because the concerns of people about housing price, a lot of researches have beenconducted to find out the main factors that affect price

Here is a brief summary of the different and most commonly factors used to affectthe price of house

III.1 Area

Many studies showed that the floor area have a positive relationship to the price of the house (Limsombunchao, 2004) This is also similar to the price of land This is because buyers are willing to pay more for a larger space, especially the functional space The land with an area larger than meet the needs of families with many

members and those who can afford to pay for a better standard of living For example,Limsombunchao (2004) studied in the housing market in New Zealand found that adding more area to increase the value of a land is about 0.08% Bajari and Kahn (2000) reported that large land area related to the price of land

3.2 Location

Location factors to be considered in many studies Factors related to the locationidentified in relation to the entire metropolitan area Location factors easiest andmost common implementation is to measure position distance from the house to thecentre which significantly impacted on land pricing which had been proven byresearchers (such as Follain and Jimenez (1985); Bajari and Kahn (2000);

land to build house to the cost of travel Positive impact of public transport services

on land prices have been examined empirically Therefore, when it comes tocalculating a home’s value, location can be more important than even the size andcondition of the house

3.3 Security

The safety of the area in which the land as located or crime rate also plays animportant role in determining land value If the area is one that is crime riddled thenthe value will be lower (Gregory Akerman, 2009) Babawale and Adewunmi (2011)indicated that the outside factors such as security, parking- lot, the distance fromapartments to church also impacts the price of real estate It is important to theexplanation of variations in land prices are variables derived from urban theory,such as distance to the CBD, and from the amenity literature, such as a community'scrime rate, arts, and recreational opportunities (Haurin and Brasington, 1996).Austin Troy and J Morgan Grove (2008) using Hedonic analysis of property data inBaltimore, they attempted to determine whether crime rate mediates how parks arevalued by the housing market The results showed that parked proximity ispositively valued by the land market where the combined robbery and rape rates for

a neighbourhood are below a certain threshold rate but negatively valued whereabove that threshold

3.4 Social infrastructure

The price of land also depends on how far social infrastructure from the land.Infrastructure is the large scale public services or systems, services and facilities of

a country or region that are necessary for economic activity, including power andwater supplies, public transportation, telecommunications, roads andschool.Closing to shopping area or shopping centre showed the impact on thevalue of surrounding residential properties Leong et al (2002) noted that there is

a shopping centre within 2 km radius making the price of land will increase byaround 0.11% in Penang, Malaysia Besides that, external benefits, includingbeautiful scenery, quiet atmosphere and the presence of urban green space hasbeen studied experimentally by Sander and Polasky (2009) used data in the city ofRamsey, United States Results also showed that people appreciated residentialareas with green space and access to the recreation area with trees The quality ofenvironment also influences prices of apartments in Brazil The apartments locatednear sewage treatment factory has low value, while near the public serviceestablishment has positive impact to the apartment’s price (Furtado 2009) All inall, real estate has no value if it has no utility, if it is not scarce and if it is noteffectively demanded In conclusion, social infrastructure has vital position when

it comes to housing price

Although the factors above are the most precise factors that affect housing priceworldwide but there is no research focusing on Viet Nam, especially the housingprice in the center city such as Ha Noi and Ho Chi Minh City So, that reallymotivated us to conduct this research and find out the result

4.1 How did we collect the data?

At first, we had to answer 4 questions: Which data to collect? (What is the mainfactors that affect the housing price), How to collect data? (Online and offline),Whose data will we collect from? (People from different ages), When to collectdata? (In Ha Noi and Ho Chi Minh City of 13 days from 17/9/2019)

Next, we listed 5 main factors that affect housing price(Age of the house, square ofthe house, the distance from the house to university, whether the house has pool ornot and whether the house has fireplace or not ) in general By listing these factors,

we can prepare the detail questions for the form It took almost 3 days to fullydeveloped the question list and make sure that all the questions were carefullychosen, also with the capable answers

We created a simple form by Google Forms and printed it out, made it asconvenient as possible for people who filled that form Most of the questions weremultiple-choice or short answers that would make people comfortable to fill itbecause it didn’t waste lots of time

All members of the group were responsible for sharing the form to as manypeople as possible that would be more and more people filling the form Bypromoting on many Facebook groups and asking people from different ages in HaNoi and Ho Chi Minh City, we finally got a quite good result: 1000 filled formswhich means 1000 observations, more than 60% forms are filled by adults indifferent universities and university students in the age group of 18 to 22 made upthe rest of the number

4.2 Explaining variables

Name

Unit of Measu re

Predictions

Dependent

variable (Y) Price Dollars

An index to evaluate price of a house

Dollars is a convertible currency, which is commonly used in a lot of transactions

The greater amount of square the house takes

up, the more creating positive effect on its price

Number of years for which the house hasbeen used (Age of house)

The longer the house has been used, the less its price would be

if the house is located close to a

If Utown = 1 means that the house is located close to at least

If not, Fplace = 0

4.3 Using GRETL to analyze the variables

4.3.1 Summary of all Variables

4.3.2 Correlation between dependent variable and each independent variables

Looking at the table above, we have some comments:

 r (Price, Sqft) = 0.5947 => low correlation, positive correlation

 r (Price, Age) = -0.0799 => very low correlation, negative correlation

 r (Price, Utown) = 0.7287 => quite high correlation, negative correlation

 r (Price, Pool)= 0.0519 => very low correlation, positive correlation

 r (Price, Fplace)= 0.0648 => very low correlation, positive correlation

V.1 Population regression function (PRF)

PRF: Price = β0 + β1× Sqft+β2×Age+β3×Utown+ β4×Pool + β5×Fplace + ui

5.2 Sample regression function (SRF)

PRF: Price = β0 + β1× Sqft+β2×Age+β3×Utown+ β4×Pool + β5×Fplace

5.3.Expectation about the variables

Dependent

variable (Y) Price

An index to evaluate price of a house

Dollars is a convertible currency, which is commonly used in a lot of transactions

highly-Independent

variables (X)

Sqft

The total square of the house (Square footage of house)

Age

Number of years for which the house has been used (Age of house)

-The longer the house has been used, the less its price would be

Utown

An index to evaluate if the house is located close to a university

or not

If Utown = 1 means that the house is located close to at leastone university

If not, Utown = 0Pool An index to

evaluate if the house

If Pool = 1 means that the house has at least 1pool

has at least

1 pool or not

If not, Pool = 0

Fplace

An index to

evaluate if the house has at least one

fireplace or not

If Fplace = 1 means thatthe house has at least 1pool

If not, Fplace = 0

VI RESULT ESTIMATION

6.1.Run the Model with GRETL:

6.2.Interpretation of the GRETL model

Number of

observation

Using observations= 1-1000

There are 1000 observations

of freedom

Co-efficient

(Coef.)

β1 = 83.1832 Holding that other factors

remain constant, when the square of house increases by 1feet, its Price increases by 83.1832 Dollars

β2 = -192.991 Holding that other factors

remain constant, when the age

of the house increases by 1 year, its Price decreases by 192.991 Dollars

β3 = 60196.2 Holding that other factors

remain constant, when there is

at least 1 university that the house is located close to, its Price increases by 60196.2 Dollars

β4 = 4352.57 Holding that other factors

remain constant, when the house has at least 1 pool, its Price increases by 4352.57 Dollars

β5 = 1398.81 Holding that other factors

remain constant, when the house has at least 1 Fireplace, its Price increases by 1398.81 Dollars

Constant

(_cons)

β0 = 6911.88 When other independent

variables equal 0, the expectedvalue of Price is 6911.88

R-squared R2= 0,8686 Indicates that the model is able to

explain 86.86% changes in the Price of the house

6.3.Rebustness Check

6.3.1.Multicollinearity Test

*Conclusion: The value of VIF of these variables are less than 10 As a result, it indicates that the model does not have multicollinearity.

6.3.2.Normality Test

JB (Jarque - Bera Test)

Hypothesis { H0:Normally Distributed

H1: Not Normally Distributed

*Conclusion: We have p-value equals to 0.934936, which is comparatively high; thus we have enough evidence not to reject H0, the model has a normal distribution.