Analysis and Discussions
The house prices in the Australian Market are different in different Australian cities. There might be various causes behind this difference. One such difference might be the type of people residing in the area. If the people residing belong to a higher class and standard in the society, the market value of the houses in that particular area will be higher than the locality in which the people residing are sub-standard. Moreover, the price index of the houses in that locality, the size of the land of the houses and also the age of the houses are also found to be significant factors to affect the market value of the house prices.
This paper is aimed at estimating the prediction model with the factors that significantly affect the market value of the houses. In order to perform the model development, data has been collected on the price index of the house price of Sydney, Melbourne and Brisbane, the percentage change in the house prices from the previous years have also been collected. All the data has been collected for a period of 14 years, starting from 2002 to 2016. Analysis has been performed in this paper only on the market value of the prices in Sydney. Thus, the age and the land size of the houses in Sydney across the fourteen years have been collected.
The value that has to be predicted in this paper is the market value of the houses. The market value of the houses is expressed in thousands of dollars. Four main variables have been considered in this research for the prediction of the market value of the houses. These are the Sydney price index of the houses, the percentage change in the house prices from the previous years, the size of the land on which the house is built and the age of the house. It has been observed from various researches that based on the price index
the market value of a house changes. Increase in the price index indicates increase in the market value. Further, the bigger the land size of the house is, the higher is supposed to be its market value. On the other hand, the more the age of a house, the less will be its market value. This is mainly because an older house needs a lot of repairing to be done before it is fit for living. All these extra expenses decrease the market value of an old house (Valadkhani & Smyth, 2017). The analysis for the model to predict the market value of the houses is discussed in the following sections.
Analysis and Discussions
Relationship between Dependent and Independent Variables
This research is to be performed in order to predict the market value of the houses in the Sydney market. Thus, the market value of houses ($000) is the dependent variable in this research. On the other hand, the market value will be predicted with the help of the information on the Sydney house price index, the percentage change in house price, the land size of the houses and the age of the houses. Thus, these four are the independent variables for the model. Relationship between each of the independent variable with the dependent variable will be illustrated graphically in this section. The relationship will be illustrated with the help of scatterplots, as it is the best graphical representation to show the association between two variables (Greenacre, 2017).
Relationship between Dependent and Independent Variables
establishes the relationship between the market value of the houses and the Sydney price index of the houses. It can be seen clearly from the figure that the price index has a positive relationship with the market value of the houses, indicating the increase in the market value with the increase in the house price index.
establishes the relationship between the market value of the houses and the percentage change in the prices of the houses from previous years. It can be seen clearly from the figure that the percentage change in the prices of the houses has a positive relationship with the market value of the houses, indicating the increase in the market value with the increase in the percentage change in the prices, though the relationship is very weak. There is very little increase in the market value of the houses with the increase in the percentage change in the prices of the houses.
establishes the relationship between the market value of the houses and the land size of the houses given in square meters. It can be seen clearly from the figure that the land size of the houses has a positive relationship with the market value of the houses, indicating the increase in the market value with the increase in the land size, though the relationship is very weak. There is very little increase in the market value of the houses with the increase in the land size of the houses.
establishes the relationship between the market value of the houses and the age of the houses given in years. It can be seen clearly from the figure that the age of the houses has a negative relationship with the market value of the houses, indicating the decrease in the market value with the increase in the age, though the relationship is very weak. There is very little decrease in the market value of the houses with the increase in the age of the houses.
To conduct the research, the following hypothesis statements are required to be framed. These are the null hypothesis (H0) and the alternate hypothesis (H1).
H0: There is no existence of a linear relationship between the market value with house price index, percentage change in price, land size and age of houses.
H1: There is an existence of a linear relationship between the market value with house price index, percentage change in price, land size and age of houses.
Regression analysis has been conducted with market value as the dependent variable and house price index, percentage change in price, land size and age of houses as the independent variables. The results of the analysis are tabulated in tables 1.1, 1.2 and 1.3. Table 1.1 indicates the summary of the regression analysis results. From table 1.1, it can be seen that the value of R-Square has been obtained as 0.79, which indicates that the percentage of variability that the dependent variables can explain is 79 percent (Darlington & Hayes, 2016).
Overall Regression Model
Table 1.2 indicates the ANOVA table obtained from the regression analysis. It can be seen from that table that the significance F value has been obtained as 0.002, which is less than the level of significance (0.05). This indicates the rejection of the null hypothesis and it can be said that linear relationship exists between the independent and the dependent variables. The model developed is thus significant overall (Schroeder, Sjoquist & Stephan, 2016).
Table 1.3 gives the results of the regression coefficients and their respective significances and confidence intervals. From the table, the following prediction equation can be estimated:
Market Value = 548.98 + (1.96 * Sydney Price Index) – (5.62 * Annual percentage change) + (0.52 * Land Size) – (2.49 * Age).
It can be observed from the coefficients that Sydney price index has positive effect on the market value, Annual percentage change has negative impact, Land size has positive effect and age has negative effect on the dependent variable market value of the Sydney houses.
Moreover, by exploring the significance values from the table, it can be seen that only the Sydney price index has a significance value less than 0.05, which indicates that in the developed model, the only significant variable is the price index of the houses. All the other variables are redundant in the model.
The confidence interval of the coefficients are also given at the confidence of 95 percent. It can be said from the obtained results with 95 percent confidence that with $1 increase in the Sydney house price index, market value increases within $660 and $3,260, with 1 percent increase in the Annual house price changes, market value can lie between decreasing $12,840 to increasing $1,600, with 1 sq. m. increase in the land size of the houses, market value can lie between decreasing $200 to increasing $1,240 and with 1 year increase in the Annual house price changes, market value can lie between increasing $5,010 to decreasing $30.
|
Table 1.1: Regression Summary Model 1 |
|
|
Multiple R |
0.89 |
|
R Square |
0.79 |
|
Adjusted R Square |
0.71 |
|
Standard Error |
43.89 |
|
Observations |
15 |
|
Table 1.2: ANOVA Model 1 |
|||||
|
|
df |
SS |
MS |
F |
Significance F |
|
Regression |
4 |
72728.59 |
18182.15 |
9.440 |
0.002 |
|
Residual |
10 |
19261.41 |
1926.14 |
||
|
Total |
14 |
91990 |
Table 1.3: Coefficients of Regression for Model 1
|
|
Coefficients |
Standard Error |
t Stat |
P-value |
Lower 95% |
Upper 95% |
|
Intercept |
548.98 |
81.13 |
6.767 |
0.000 |
368.21 |
729.75 |
|
Sydney price Index |
1.96 |
0.58 |
3.367 |
0.007 |
0.66 |
3.26 |
|
Annual % change |
-5.62 |
3.24 |
-1.735 |
0.113 |
-12.84 |
1.60 |
|
Total number of square meters |
0.52 |
0.32 |
1.603 |
0.140 |
-0.20 |
1.24 |
|
Age of house (years) |
-2.49 |
1.13 |
-2.202 |
0.052 |
-5.01 |
0.03 |
Regression Analysis with Land Size
To conduct the research, the following hypothesis statements are required to be framed. These are the null hypothesis (H01) and the alternate hypothesis (H11).
H01: There is no existence of a linear relationship between the market value and land size of the houses.
H11: There is an existence of a linear relationship between the market value and land size of houses.
Regression analysis has been conducted with market value as the dependent variable and land size as the independent variable. The results of the analysis are tabulated in tables 1.4, 1.5 and 1.6. Table 1.4 indicates the summary of the regression analysis results. From table 1.1, it can be seen that the value of R-Square has been obtained as 0.10, which indicates that the percentage of variability that the dependent variables can explain is 10 percent.
Table 1.5 indicates the ANOVA table obtained from the regression analysis. It can be seen from that table that the significance F value has been obtained as 0.256, which is higher than the level of significance (0.05). This indicates the acceptance of the null hypothesis and it can be said that linear relationship does not exist between the independent and the dependent variable. The model developed is thus insignificant.
Table 1.3 gives the results of the regression coefficients and their respective significances and confidence intervals. From the table, the following prediction equation can be estimated:
Market Value = 659.14 + (0.56 * Total number of square meters).
|
Table 1.4: Regression Summary Model 2 |
|
|
Multiple R |
0.31 |
|
R Square |
0.10 |
|
Adjusted R Square |
0.03 |
|
Standard Error |
79.89 |
|
Observations |
15 |
|
Table 1.5: ANOVA Model 2 |
|||||
|
|
df |
SS |
MS |
F |
Significance F |
|
Regression |
1 |
9026.56 |
9026.56 |
1.414 |
0.256 |
|
Residual |
13 |
82963.44 |
6381.80 |
||
|
Total |
14 |
91990 |
Table 1.6: Coefficients of Regression for Model 2
|
|
Coefficients |
Standard Error |
t Stat |
P-value |
Lower 95% |
Upper 95% |
|
Intercept |
659.14 |
101.22 |
6.512 |
0.000 |
440.47 |
877.82 |
|
Total number of square meters |
0.56 |
0.47 |
1.189 |
0.256 |
-0.46 |
1.59 |
However, by comparing the two models developed, it can be stated without a doubt that the overall model developed at first is a much better prediction model than the model with only land size as the independent variable. As can be seen that the accuracy in prediction for the first model is 79 percent whereas the accuracy in prediction for the second model is only 10 percent. Moreover, the second model has been found to be insignificant. Thus, in any case, the first model developed considering all the four independent variables is a much better model for prediction of the market value of the Sydney houses.
According the second equation developed, the market value of a house in Sydney with a land size of 400 sq. m. is given by:
Market Value ($000) = 659.14 + (0.56 * 400) = 883.14.
Thus, the required market value of a house with 400 sq. m area is $883,140 in Sydney.
Conclusion
In this research the market value of the houses in Sydney has to be predicted. The prediction has been conducted with the help of regression analysis. Data has been collected on 14 years from 2002 to 2016. The Sydney price index of houses, Annual percentage change in house prices, land size and age of the houses are considered as the predictors. From the model, Annual percentage change in house prices, land size and age have been found to be insignificant and Sydney price index of houses is significant for the prediction of the market values.
References
Darlington, R. B., & Hayes, A. F. (2016). Regression analysis and linear models: Concepts, applications, and implementation. Guilford Publications.
Greenacre, M. (2017). Correspondence analysis in practice. Chapman and Hall/CRC.
Schroeder, L. D., Sjoquist, D. L., & Stephan, P. E. (2016). Understanding regression analysis: An introductory guide (Vol. 57). Sage Publications.
Valadkhani, A., & Smyth, R. (2017). Self-exciting effects of house prices on unit prices in Australian capital cities. Urban Studies, 54(10), 2376-2394.
Order an Essay Now & Get These Features For Free:
Turnitin Report
Formatting
Title Page
Citation
Outline
