## ← 03_s - Intro to Machine Learning

• 1 Follower
• 29 Lines

### Get Embed Code x Embed video Use the following code to embed this video. See our usage guide for more details on embedding. Paste this in your document somewhere (closest to the closing body tag is preferable): ```<script type="text/javascript" src='https://amara.org/embedder-iframe'></script> ``` Paste this inside your HTML body, where you want to include the widget: ```<div class="amara-embed" data-url="http://www.youtube.com/watch?v=61sZUjEPzt0" data-team="udacity"></div> ``` 5 Languages

Showing Revision 1 created 10/10/2016 by Udacity Robot.

1. So to solve this, we do the same trick as before.
2. And again I lined it up at the number so
3. correctly, it's not as easy to do real regression.
4. But, you're going to find that as we increase our size by 500 feet
5. from 500 to 1000, or 1000 to 1500, the price also increases by 500.
6. So that means the ratio of price increased to size equals one.
7. And that goes in over here.
8. So go vertical, we find that if we increase the age by 20 years,
9. our price drops by minus 200.
10. So that means the ratio of those minus 200 that's divided by 20 is minus 10.,
11. It's really important to get the minus sign right,
12. because older houses are cheaper in this spreadsheet.
13. And eventually as we try and find the y intercept so to speak.
14. Or put differently the constant over here.
15. We plug in those two constants over here on the left data point.
16. 500 times one would be 500.
17. Zero times minus ten is zero.
18. So we would with these two coefficients only we get a 500 over here.
19. If we need to get all the way to 1,000, so we have to add 500 to it.
20. And that's the correct result.
21. Okay once again this is a multivariate regression.
22. It happens to be an exact fit.
23. If you get into situations where these numbers don't exactly fit,
24. you have to apply a much more complicated formula.
25. You can't just look at the data over here and data over here.
26. You have to look at all the data, but
27. explaining the detailed formula to you is a bit outside the scope of this class.
28. But what you will do is, we'll use the existing software
29. to calculate results for situations just like this