In this section, we will apply least square problems to economics.

Instead of finding the least squares solution of

$Ax=b$, we will be finding it for

$X\beta =y$ where

$X$→ design matrix

$\beta$→ parameter vector

$y$→ observation vector

**Least-Squares Line**
Suppose we are given data points, and we want to find a line that best fits the data points. Let the best fit line be the linear equation

$y=\beta_0 + \beta_1 x$
And let the data points be

$(x_1,y_1 ),(x_2,y_2 ),\cdots,,(x_n,y_n )$. The graph should look something like this:

Our goal is to determine the parameters

$\beta_0$ and

$\beta_1$. Let’s say that each data point is on the line. Then

This is a linear system which we can write this as:

Then the least squares solution to

$X\beta=y$ will be

$X^T X \beta =X^T y$.

__General Linear Model__
Since the data points are not actually on the line, then there are residual values. Those are also known as errors. So we introduce a vector called the residual vector

$\epsilon$, where

$\epsilon = y - X\beta$

→$y = X\beta + \epsilon$
Our goal is to minimize the length of

$\epsilon$ (the error), so that

$X\beta$ is approximately equal to

$y$. This means we are finding a least-squares solution of

$y=X\beta$ using

$X^T X\beta=X^T y$.

**Least-Squares of Other Curves**
Let the data points be

$(x_1,y_1 ),(x_2,y_2 ),\cdots,,(x_n,y_n )$ and we want to find the best fit using the function

$y=\beta_0+\beta_1 x+\beta_2 x^2$, where

$\beta_0,\beta_1,\beta_2$ are parameters. Technically we are using a best fit quadratic function instead of a line now.

Again, the data points don’t actually lie on the function, so we add residue values

$\epsilon_1,\epsilon_2,\cdots,\epsilon_n$ where

Since we are minimizing the length of

$\epsilon$, then we can find the least-squares solution

$\beta$ using

$X^T X\beta=X^T y$. This can also be applied to other functions.

__Multiple Regression__
Let the data points be

$(u_1,v_1,y_1 ),(u_2,v_2,y_2 ),\cdots,,(u_n,v_n,y_n )$ and we want to use the best fit function

$y=\beta_0+\beta_1 u+\beta_2 v$, where

$\beta_0,\beta_1,\beta_2$ are parameters.

Again, the data points don’t actually lie on the function, so we add residue values

$\epsilon_1,\epsilon_2,\cdots,\epsilon_n$ where

Since we are minimizing the length of

$\epsilon$, then we can find the least-squares solution

$\beta$ using

$X^T X\beta=X^T y$. This can also be applied to other multi-variable functions.

1.

**Applications to Linear Models Overview:**

a)

__Applying Least-Squares Problem to Economics__

• Go from $Ax=b$ to $X\beta=y$

• $X$→ design matrix

• $\beta$→ parameter vector

• $y$→ observation vector

b)

__Least-Squares Line__

• Finding the best fit line

• Turning a system of equations into $X\beta =y$

• Using the normal equation $X^T X\beta=X^T y$

• Introduction of the residual vector

c)

__Least-Squares to Other Curves__

• Finding the Best Fit Curve (not a line)

• Using the normal equation $X^T X\beta=X^T y$

d)

__Least-Squares to Multiple Regressions__

• Multiple Regression → multivariable function

• Finding a Best Fit Plane

• Using the normal equation $X^T X\beta=X^T y$

2.

**Finding the Least-Squares Line**

Find the equation $y=\beta_0+\beta_1 x$ of the least-squares line that best fits the given data points:

$(0,1),(1,2),(2,3),(3,3)$

3.

**Finding the Least-Squares of Other Curves**

Suppose the monthly costs of a product depend on seasonal fluctuations. A curve that approximates the cost is

$y= \beta _0 + \beta _1 x+ \beta _2 x^2 + \beta_3 \cos$ ($\frac{2 \pi x}{12}$)

Suppose you want to find a better approximation in the future by evaluating the residual errors in each data point. Let’s assume the errors for each data point to be $\epsilon_1,\epsilon_2,\cdots,\epsilon_n$.

Give the design matrix, parameter vector, and residual vector for the model that leads to a least-squares fit for the equation above. Assume the data are $(x_1,y_1 ),\cdots,(x_n,y_n).$

4.

An experiment gives the data points $(0,1) , (1,3) , (2, 4), (3, 5)$. Suppose we wish to approximate the data using the equation

$y=A+Bx^2$

First find the design matrix, observational vector, and unknown parameter vector. No need to find the residual vector. Then find the least-squares curve for the data.

5.

**Finding the Least Squares of Multiple Regressions**

When examining a local model of terrain, we examine the data points to be $(1,1, 3), (2, 2, 5),$ and $(3, 1, 3)$. Suppose we wish to approximate the data using the equation

$y=\beta_0 u+\beta_1 v$

First find the design matrix, observational vector, and unknown parameter vector. No need to find the residual vector. Then find the least-squares curve for the data.

6.

**Proof Question Relating to Linear Models**

Show that

$\lVert X \hat{\beta} \rVert^2=\beta^TX^Ty$