ECON4003: Practical Assignment Part I (a) π₯π − π = πΌππ E π₯π − π ππ = πΌ π₯π ππ ππ = − πΌ πΌ π₯π ππ − = π½π + π½1 π₯π + π’π πΌ πΌ π₯π ππ π’π = − − π½0 − π½1 π₯π πΌ πΌ 1 ππ = ( − π½1 ) π₯π − − π½0 πΌ πΌ 1 − πΌπ½1 ππ =( ) π₯π − − π½0 πΌ πΌ 1 − πΌπ½1 ππ = π₯π − π½0 − πΌ πΌ 1−πΌπ½1 (b) πΈ(π’π |π) = πΈ ( πΌ π ππ − π½0 − πΌπ |π) 1 − πΌπ½ π = πΈ ( πΌ 1 ππ |π) − πΈ(π½0|π) − πΈ ( πΌπ |π) 1 − πΌπ½1 1 (πΈ(ππ |π)) − πΈ(π½0 |π) − πΈ(ππ |π) πΌ πΌ 1 − πΌπ½1 1 = ππ − π½0 − × 0 πΌ πΌ 1 − πΌπ½1 = ππ − π½0 πΌ = 1 (c) Assumption SR.3 confirms that the sample outcomes of ππ , π = 1, … , π , must take at least two different values π Μ1 = ∑ π€π ππ π½ π=1 π = ∑ π€π (π½0 + π½1 ππ + π’π ) π=1 This substitution is possible by inputting the true model expression derived from the linearity assumption π π π π½Μ1 = π½0 ∑ π€π + π½1 ∑ π€π ππ + ∑ π€π π’π π=1 π=1 π=1 π = π½1 + ∑ π€π π’π π=1 since π ∑ π€ = 0, π=1 π ∑ π€ππ = 1 π=1 Taking the expectations of π½Μ1 conditional on the sample values of regressor X, where π€π is treated as non-random in this case since it is a function only of X π πΈ(π½Μ1|π) = πΈ(π½1 + ∑ π€π π’π |π) π=1 π = πΈ(π½1 |π) + πΈ(∑ π€π π’π |π) π=1 π = π½1 + ∑ π€π πΈ(π’π |π) π=1 2 π½1 is a constant and thus is its own expected value and π€π can be removed from the conditional expectation expression as it is non random 3 (d) 4