Assignment 1

Non-Linear Regression Analysis

n
X

1. Suppose {x1 , . . . , xn } is a set of n real numbers. Find a such that

(xi a)2 is

i=1

minimized.
2. Suppose {x1 , . . . , xn } is a set of n real numbers. Find a such that

n
X

|xi a| is

i=1

minimized.
3. Suppose {(x1 , y1 ), . . . , (xn , yn )} is a set of n pairs. Find a such that

n
X

|yi axi | is

i=1

minimized.
4. Consider the following simple linear regression model:
yi = a + bxi + i ;

i = 1, . . . , n.

Here i s are i.i.d. normal random variables with mean 0 and variance 1. (a) Find ab and
the maximum likelihood estimators of a and b. (b) Are ab and bb always consistent?
Find a sufficient condition under which they are consistent.

b
b,

5. Consider the same problem as in Problem 4. But it is known that 0 a 1 and

0 b 1. Answer (a) and (b).
6. Consider the following simple linear regression model
yi = bxi + i ;

i = 1, . . . , n.

Here i s are i.i.d. Laplace random variables with mean 0 and variance 1. Find the
maximum likelihood estimator of b. Find a sufficient condition so that it is consistent.
7. Consider the following simple linear regression model:
yi = a + bxi + i ;

i = 1, . . . , n.

Here i s are Laplace random variables with mean 0 and variance 1, it is known that
0 a 1 and 0 b 1. Let us assume you have enough computer power in your
disposal. Can you suggest a method to compute the maximum likelihood estimators
of a and b.
8. Consider the following simple linear regression model:
yi = a + bxi + i ;

i = 1, . . . , n.

Here i s are Laplace random variables with mean 0 and variance 1. It is known that
a = 2b, find the maximum likelihood estimator of a.