Chapter I
Introduction
1.1 Background of the study
Researches add value to us. Likewise, this study is done to relate the practical decision making on daily life on the basis of statistical information. When we walk from New-Baneshwor to Old-Baneshwor we see number of hostel serving either boy or girl. They are serving especially the student group. They take certain remuneration for the service. A desire arises within us, which variables actually affect the fee structure of hostel. Then we decided to make a small research on the hostel rent in our context.
1.2 Objective of the study
Main objectives of the following study are as follows:
To know variables that affect the hostel fee structure
To be practically familiar with statistical tool, SPSS
How different variables such as number of person in a room, furnished room, floor of the room, etc affect the rent of hostel
1.3 Limitations of the study
Information is collected from hostel between Old- Baneshwor and New- Baneshwor.
Few variables are made into consideration.
So the conclusion drawn from this research finding cannot be equally generalized in other settings.
1.4 Research Methodology
Research methodology is primarily empirical type. The data are collected primarily from hostel between Old- Baneshwor and New- Baneshwor. In this report, we use Descriptive and Analytical research design.
A simple questionnaire was used to collect information. We individually visited six hostels. Information is collected from respective owner of hostel. There are total 287 boys or girls living in these hostels.
Chapter II
Data Analysis
2.1 Tools for Analysis
- Regression:
It is a statistical measure that attempts to determine the strength of the relationship between one dependent variable (usually denoted by Y) and a series of other changing variables (known as independent variables).
The two basic types of regression are linear regression and multiple regressions. Linear regression uses one independent variable to explain and/or predict the outcome of Y, while multiple regressions use two or more independent variables to predict the outcome.
- Multi Collinearity:
Multi co linearity is a statistical phenomenon in which two or more predictor variables in a multiple regression model are highly correlated. In this situation the coefficient estimates may change erratically in response to small changes in the model or the data.
A high degree of multi co linearity can also cause computer software packages to be unable to perform the matrix inversion that is required for computing the regression coefficients, or it may make the results of that inversion inaccurate. The degree of multicollinearity is determined by a test called VIF. If VIF<10, there exists no multicollinearity, and if VIF>10, there exists multicollinearity.
- ANOVA:
In statistics, analysis of variance (ANOVA) is a collection of statistical models, and their associated procedures, in which the observed variance in a particular variable is partitioned into components attributable to different sources of variation. In its simplest form ANOVA provides a statistical test of whether or not the means of several groups are all equal, and therefore generalizes t-test to more than two groups. Doing multiple two-sample t-tests would result in an increased chance of committing a type I error. For this reason, ANOVAs are useful in comparing two, three or more means.
- Autocorrelation:
Autocorrelation is a statistical tool which is used to find the effect of seasonal data in the statistics. In many of the cases, the seasonal data may affect the calculation and interpretation of the data. The effect is calculated by Durban Waston test, simply known as D-W test. The variables could have either auto correlation, no auto correlation of inconclusive. Factors like dL and dU are used to determine the value of D-W test.
2.2 Analysis and Findings
After entering the data in the SPSS model, we came to analyze the following findings with the above mentioned tools.
- Correlation:
Correlation coefficient(r) =0.670
which implies that there is positive correlation between dependent variable (Hostel rent) and independent variables (no. of persons in a room, floor, sex, furnished).
Coefficient of Determination:
R^2=0.449
Which implies 44.9% of the variation in hostel rent is explained by this regression equation.
Standard Error of Estimate:
Standard error of estimate
Syx=793.68
Which means the variability of observed value of hostel rent from the regression line is Rs 793.68.
- Regression:
The regression equation that explains the relationship between dependent variable(Y) and dependent variables(X1, X2 and X3) is given by:
Y=7392.456-208.201X1-584.565X2+136.518X3
Where Y¬¬-hostel rent
X1-No of floor
X2-No of persons in a room
X3-Gender (i.e. 0=Female; 1=Male)
Bo=7392.46
This implies other factors rather than independent variables (mentioned above) th
at determine hostel rent.
B1=-208.201
This implies increase in floor by one unit leads to decrease in price of hostel rent by Rs. 208.20 keeping other variables constant.
B2= -584.565
This implies that increase in no. of persons in a room by one unit leads to decrease in price of hostel rent by Rs. 584.57 keeping other factors constant.
B3= 136.518
This implies that the hostel rent for male is greater than that of female by Rs. 136.52 keeping other factors constant.
Test of Regression Coefficient (Slope):
Floor
H0: B1=0 i.e. there is no linear relationship between hostel rent and level of floor.
H1: B1≠0 i.e. There is significant relationship between hostel rent and the level of floor.
P-value = 0.417
Decision:
At 0.05 level of significance, there exists no relationship between hostel rent and level of floor.
Number of persons in a single room
H0: B2=0 i.e. there is no linear relationship between hostel rent and no. of persons in a room
H1: B2≠0 i.e. there is linear relationship between hostel rent and no. of persons in a room
P-value = 0.004
Decision:
At 0.05 level of significance, there exists relationship between hostel rent and no. of persons in a room
Gender
H0: B3=0 i.e. there is no linear relationship between hostel rent and gender
H1: B3≠0 i.e. there is linear relationship between hostel rent and gender
P-value = 0.775
Decision:
At 0.05 level of significance, there exists no relationship between hostel rent and gender.
Test of Significance as a whole
H0: B1=B2=B3=0 i.e. Regression is not significant as a whole
H1: At least one Bi≠0 i.e. Regression is significant as a whole
P-value = 0.020
Decision:
At 0.05 level of significance, the regression is significant as a whole.
Test of multicollinearity
Since all VIFs are less than 10, there exists no multicollinearity between independent variables.
Test of autocorrelation
We can’t test autocorrelation because it is cross sectional data.
Chapter: 3
Conclusion
From the above calculation and we obtained the linear relationship between Hoste Rent and Number of persons in a single room. At 0.05 level of significance, regression is significant as a whole. Likewise there exists no multicollinearity between independent variables.
Since coefficient of determination (R-square) is 0.449, there seems lack of other independent variables.
Thank You!
No comments:
Post a Comment