1. Consider the model Y_{t} = β_{0} + β_{1} X_{t} + ε_{t}, where t = 1,..., n. If the errors ε_{t} are not correlated, then the OLS estimates of β_{0} and β_{1} will be unbiased.
2. In the following regression model ln Y_{i }= β_{0} + β_{1} X_{1i} + β_{2} ln X_{2i} + ε_{i}, all β_{k} coefficients measure the elasticities of the Y variable with the respective X variables, because the Y variable appears in a logarithmic form.
3. If we want to capture a curvature in the relationship between Y_{i} and X_{i}, we have to use a quadratic model, where the slope is not constant everywhere and changes according to the value of Xi at which it is being assessed.
4. If a hypothesis is rejected at the 0.10 level of significance, it may not be rejected at the 0.05 level of significance.
5. If β_{1} is positive in the equation
(1) Y_{i} = β_{0} + β_{1} X_{1i} + β_{2} X_{2i} + ε_{i}, it can never be negative in the equation
(2) Y_{i} = β_{0} + β_{1} X_{1i} + β_{2} X_{2i} + β_{3} X_{3i} + ε_{i},.