Exam 1

  1. Question

    Consider the following regression results:
    Call:
    lm(formula = log(y) ~ log(x), data = d)
    Residuals:
        Min      1Q  Median      3Q     Max 
    -6.6119 -1.4477  0.1735  1.5365  4.8160 
    Coefficients:
                Estimate Std. Error t value Pr(>|t|)
    (Intercept)   0.1264     0.2520   0.501    0.618
    log(x)        0.2870     0.2279   1.259    0.212
    Residual standard error: 2.251 on 79 degrees of freedom
    Multiple R-squared:  0.01967,	Adjusted R-squared:  0.007263 
    F-statistic: 1.585 on 1 and 79 DF,  p-value: 0.2117
    
    Describe how the response y depends on the regressor x.

    Solution

    The presented results describe a log-log regression.
    The mean of the response y increases with increasing x.
    If x increases by 1 percent then a change of y by about 0.29 percent can be expected.
    However, the effect of x is not significant at the 5 percent level.