代写英文：家庭娱乐支出的因素和变量
这些变量包括与受访者的地域和地理位置相对应的变量。属于农村地区的家庭被纳入变量“rural”，该变量对农村受访者的值为1。13.7%的受访者来自农村地区。有九个虚拟变量对应于加拿大的九个省。对于组中的k个类别，需要k-1虚拟变量来避免线性依赖(Gujarati & Sangeetha, 2003)。因为有10个省份的数据收集,9为模型假人被基地省份安大略。4.3%的受访者来自纽芬兰,4.2%属于爱德华王子岛,7.9%来自新斯科舍省,6.3%来自新布伦瑞克,14.9%受访者来自魁北克,6%来自马尼托巴省,8.6%来自萨斯喀彻温省,7.8%来自阿尔伯塔省,13.8%来自不列颠哥伦比亚省和26.3%属于安大略省。
定量分析:
这里估计的有效模型为:
Y = +我团队= 116 +
其中α是常数项的回归,βis相对应的系数估计不同变量切除酶和µ误差项。表3为模型回归结果，因变量为游憩费用日志。
系数统计显著性的假设检验:
系数统计显著性t检验背后的假设为:
H0: i=0 i=1 to 16H1: i0 i=1 to 16
计算各系数的t统计量，并与学生t分布表中的临界t值进行比较。另一种方法是检查t统计量对应的p值。如果小于0.1大于0.05,系数统计不同从0 10%水平的意义,如果假定值位于0.01和0.05之间,零假设被拒绝在5%水平的意义,如果假定值小于0.01,零假设被拒绝在1%水平的意义。反之，如果p值大于0.1，则不拒绝零假设，且系数与零无统计学差异。
例如，lnincome系数的p值为~0，即在1%的显著性水平下，该系数具有统计学意义，或与0有统计学差异，或拒绝零假设。
为了检验整体模型是否具有统计学意义，使用方差分析或方差分析。其结果如表3所示。检验的原假设为:
H0: =1=2==16=0H1:至少有一个系数与0不同
测试采用F-test。在给定的情况下，f统计量为404.235，其自由度为(16,10186)。对应的p值为~0。因此，拒绝原假设，模型在1%的显著性水平上具有统计学意义。
代写英文 ：家庭娱乐支出的因素和变量
These variables include those corresponding to the locality and geography of the respondents. The families belonging to rural areas are taken in the variable “rural” which takes a value 1 for rural respondents. 13.7% respondents are from rural areas. There are nine dummy variables corresponding to nine provinces of Canada. For k categories in a group, k-1 dummy variables are needed to avoid linear dependency (Gujarati & Sangeetha, 2003). Since there are 10 provinces for which data is collected, 9 dummies are taken for the model with the base province being Ontario. 4.3% respondents come from Newfoundland, 4.2% belong to Prince Edward Island, 7.9% come from Nova Scotia, 6.3% hail from New Brunswick, 14.9% respondents are from Quebec, 6% are from Manitoba, 8.6% are from Saskatchewan, 7.8% are from Alberta, 13.8% are from British Columbia and 26.3% belong to Ontario.
Quantitative Analysis:
The effective model estimated here is:
Y=+i=116iXi+
Where α is the constant term of the regression, βis are the coefficients estimated corresponding to different variables Xis and µ is the error term. Table 3 shows the regression results of model where the dependent variable is the log of recreational expenses.
Hypothesis testing for statistical significance of coefficients:
The hypotheses behind the t-test for statistical significance of the coefficients are:
H0: i=0 i=1 to 16H1: i0 i=1 to 16
For each coefficient, the t-statistic is calculated and compared with the critical t value from student’s t-distribution table. An alternative approach is to check the corresponding p-value of the t-statistic. If it is less than 0.1 but greater than 0.05, the coefficient is statistically different from 0 at 10% level of significance, if p-value lies between 0.01 and 0.05, the null hypothesis is rejected at 5% level of significance and if the p-value is less than 0.01, the null hypothesis is rejected at 1% level of significance. On the contrary, the null hypothesis is not rejected and the coefficient is not statistically different from zero, if the p-value is greater than 0.1.
For example, the p-value of the coefficient of lnincome is ~0 which means that the coefficient is statistically significant or the coefficient is statistically different from 0 or the null hypothesis is rejected, at 1% level of significance.
To check if the overall model is statistically significant, ANOVA or Analysis of Variance is used. Its result is shown in table 3. The null hypothesis for the test is:
H0: =1=2==16=0H1:At least one of the coefficients is different from 0
The test is done by F-test. In the given case, the F-statistic is 404.235 and its degrees of freedom are (16,10186). The corresponding p-value is ~0. Hence the null hypothesis is rejected and the model is statistically significant at 1% level of significance.
