linear regression.

So, that's linear regression.

linear regression for classification problems.

So, here's our linear regression algorithm.

linear regression isn't a good idea .

So linear regression works well if the data is approximately linear, but there are many examples when linear regression performs poorly.

So that's linear regression with gradient descent.

And thus, using linear regression this way.

Linear regression? Do we compute confidence intervals?

So outliers are particularly bad for linear regression.

Our first learning algorithm will be linear regression.

The first thing to understand is the regression equation and the different components of the regression equation.

So again, here's the simple regression equation.

You now know about linear regression and gradient descent.

And let's say you've implement and regularize linear regression.

This model is called linear regression or this, for example, is actually linear regression with one variable, with the variable being x.

That regression, or that regression equation, we could run it in R, and get the regression equation. We'll, we'll do that in, in lecture nine.

So we have our simple regression equation. Here.

And that's what regression equation is showing us.

A linear separator is a linear equation that separates positive from negative examples.

Thrun In this question I quiz you about linear regression.

Right? This was our objective function for the linear regression.

linear regression for fitting a straight line to our data.

This line over here is linear regression to the data.

With this last regression equation. M should be significant.

We'll just put one predictor in the regression equation.

And now we have a level one linear equation.

So what is a linear second order differential equation?

So now we've simplified this to a linear equation.

Let's solve another 2nd order linear homogeneous differential equation.

So that also makes this a non linear equation.

So, I'll just summarize the multiple regression equation as follows.

Those are the coefficients that go into the regression equation.

Here I've, I've summarized the regression equation at the top.

Now that we're in regression I want to present this with respect to the unstandardized regression coefficient, B, in the regression equation.

We already have linear regression and we have logistic regression, so why do we need, you know, neural networks?

So this equation right here at first it doesn't look like a straightforward linear equation

level three system, level three linear equation but the bottom

So this is just a linear equation with two unknowns.

Before, when we just did simple regression, that was our equation.

It was just that equation with one regression coefficient in it.

Here's our regression equation. We have the intercept and the slope.

So, here's a simple regression equation, or, a simple path model.

Then linear regression tries to fit a line that best describes the data.

Are there closed form solutions of the type we found for linear regression?