1
00:00:00,025 --> 00:00:04,824
Okay, so
the first regression task that we have is

2
00:00:04,824 --> 00:00:09,526
we have to figure out what
model are we gonna use.

3
00:00:09,526 --> 00:00:13,374
Are we gonna assume that there is just
a constant relationship between square

4
00:00:13,374 --> 00:00:14,390
footage and price?

5
00:00:14,390 --> 00:00:16,560
That means regardless of
the size of the room,

6
00:00:16,560 --> 00:00:19,960
we are expecting every house to sell for
the same amount.

7
00:00:20,960 --> 00:00:23,020
Well that's probably not a great model.

8
00:00:23,020 --> 00:00:26,270
Are we gonna assume that there's
some linear relationship?

9
00:00:27,580 --> 00:00:30,190
So as I increase square footage,

10
00:00:30,190 --> 00:00:34,640
my price increases at the same rate
as I'm increasing square footage.

11
00:00:34,640 --> 00:00:37,470
Or I'm I gonna assume that
there's some quadratic fit or

12
00:00:37,470 --> 00:00:41,470
some higher order polynomial fit, or
the list of models I could consider

13
00:00:41,470 --> 00:00:46,170
is very long and that's what this
course is partially gonna be about.

14
00:00:46,170 --> 00:00:51,990
Exploring different options that
we have for models of our data.

15
00:00:51,990 --> 00:00:56,460
Okay, so one task is out of the space of
all these models that we might consider.

16
00:00:56,460 --> 00:01:00,960
Which is the one that we should use for
a given dataset and task that we have?

17
00:01:02,550 --> 00:01:07,301
Okay, but now, let's assume that we

18
00:01:07,301 --> 00:01:12,645
have selected the model we're gonna use,

19
00:01:12,645 --> 00:01:17,991
in this case, here, we're assuming that

20
00:01:17,991 --> 00:01:23,187
we're gonna use just a quadratic fit, so

21
00:01:23,187 --> 00:01:29,290
assume Model, f of x,
is a quadratic function.

22
00:01:35,020 --> 00:01:41,830
Then our next task is gonna be to estimate
a specific quadratic fit to the data.

23
00:01:42,830 --> 00:01:46,210
Okay, so a model just specifies
the form of something,

24
00:01:46,210 --> 00:01:49,360
it's gonna be defined in terms
of some set of parameters.

25
00:01:49,360 --> 00:01:55,270
And then we're gonna have to estimate
what the specific fit is from the data.

26
00:01:55,270 --> 00:01:59,651
So for example, here,

27
00:01:59,651 --> 00:02:06,666
this is our estimated quadratic fit.

28
00:02:09,892 --> 00:02:14,349
And we'll call it f hat of x,
this is our estimated

29
00:02:14,349 --> 00:02:18,820
function that's fit from
our specific dataset.

30
00:02:21,810 --> 00:02:24,660
Or this is another function
that we could have fit.

31
00:02:25,930 --> 00:02:29,460
And we'll talk about the way
in which we're gonna fit

32
00:02:31,220 --> 00:02:34,590
functions to data in this course.

33
00:02:34,590 --> 00:02:39,241
Okay, but the point is that first we
have to choose a model, then we have to

34
00:02:39,241 --> 00:02:44,290
provide some procedure, some algorithm,
for fitting that model to the data.

35
00:02:44,290 --> 00:02:47,523
And coming up with a specific
curve that we're gonna use for

36
00:02:47,523 --> 00:02:49,211
our tasks such as prediction.

37
00:02:49,211 --> 00:02:53,379
[MUSIC]