This report tries to analyse the factors and variables that determine and influence the recreational expenditure of the households in Canada. Recreational expenditures are important because not only do they improve the quality of life of the residents, but also boost tourism in the economy.
The basic equation to follow the econometric model for the purpose here is:
Recreational Expenditure = f (Individual Characters, Social Characters)
Individual Characters refer to those characteristics that pertain to an individual like his or her gender, education, age etc.
Social Characters refer to those characteristics that correspond to the whole household like the household income, geographical location and so on.
The explanatory power of the model is 38.8% and the model is significant at 1% level of significance. The presence of multicollinearity is less likely in the model because in the presence of multicollinearity, the variables lose their statistical significance and the explanatory power of the model increases very strongly (Greene, 2000). This however does not hold for this model. The model can be heteroskedastic because of specification error. The presence of heteroskedasticity is often found due to specification error. It is evident from figure 1 that a linear model might not be a good specification for the dependent variable.
The coefficient of log of income is 1.119 and is statistically significant at 1% level of significance. This means that for a unit increase in the log of income, the log of recreational expenditures in the household increases by 1.119 when all other variables are held constant. The coefficient of sex is negative implying that females are more inclined towards recreational activities than males in Canada. The coefficient of education is also negative and is significant at 5% level of significance. This implies that for a unit increase in the education, the log of recreational expenditures is likely to reduce by 0.009. The coefficient of social assistance is also negative. Since the variable takes a value of 1 for the households which receive social assistance, the negative coefficient implies that such families are less likely to engage much in recreational activities. Among the provincial dummy variables only the dummies corresponding to Quebec, Saskatchewan and British Columbia are significant. The base category here is Ontario. Since the coefficient of Quebec is negative, the people of Quebec are less likely to spend on such activities as compared to Ontario. On the other hand, people from Saskatchewan and British Columbia are more likely to do so. The coefficient of rural households is negative, implying that these households are expected to spend less on recreational activities. The coefficient of age is -0.018 which implies that for a unit increase in the age of the Canadians, the log of recreational expenses is expected to decline by 0.018.
For a unit increase in log of income, the increase in log of expenses on recreational activities is more than 1. This means that the people in Canada prefer spending on recreational activities. However, the desire is limited to certain areas. Like Rural areas do not spend much on recreational activities and so do people of certain provinces. There needs to be a program in place to stimulate such expenses. It is important because expenditure in such activities leads to a better quality of life which in turn leads to better human productivity. Since age has a negative effect on recreational expenses, dedicated recreational avenues need to be developed for people of higher ages.