1
00:00:00,360 --> 00:00:03,240
In this video, we will be learning about white noise.

2
00:00:04,680 --> 00:00:12,810
We are learning about white noise, because if the series that we want to forecast that cities is white

3
00:00:12,810 --> 00:00:16,830
noise, then we cannot predict that cities.

4
00:00:18,630 --> 00:00:27,660
We cannot forecast that RTD white noise basically means that it is a sequence of random numbers and

5
00:00:27,660 --> 00:00:30,020
we cannot forecast random numbers.

6
00:00:32,270 --> 00:00:38,330
So in this video, we're going to learn about some techniques which we can use to identify whether the

7
00:00:38,330 --> 00:00:40,550
series is white noise or not.

8
00:00:42,140 --> 00:00:44,030
We will do this for two reasons.

9
00:00:47,130 --> 00:00:53,310
Firstly, we will try to see whether our cities to be forecasted is Vytenis or not.

10
00:00:54,180 --> 00:01:02,820
If it is white noise, it cannot be predicted and we will stop the process there only if it is not white

11
00:01:02,820 --> 00:01:03,330
noise.

12
00:01:03,930 --> 00:01:08,790
We will do the predictions, find out the errors in our predictions.

13
00:01:09,540 --> 00:01:14,730
That is, we'll find out the difference between actual value and the predicted values.

14
00:01:16,020 --> 00:01:20,970
And then we will see if these error values are white noise or not.

15
00:01:23,390 --> 00:01:31,610
If these air values are not white noise, then this means that there is still information which could

16
00:01:31,610 --> 00:01:32,450
have been modelled.

17
00:01:33,470 --> 00:01:40,220
But our current model was not able to take out that information in such a case.

18
00:01:40,520 --> 00:01:43,490
We may look for some more advanced forecasting model.

19
00:01:45,320 --> 00:01:53,480
Whereas if the forecast errors are white noise, it means that all information has been hardness and

20
00:01:53,600 --> 00:01:56,660
all that is left is random fluctuations.

21
00:01:59,700 --> 00:02:00,810
So in somebody's.

22
00:02:03,560 --> 00:02:12,230
Our cities should not be white noise so that we can model it and forecast it and our errors should be

23
00:02:12,230 --> 00:02:19,370
white noise so that the model that we are using is capturing all the information in our cities.

24
00:02:21,680 --> 00:02:27,950
With this background, let us look at the tools which are used to identify whether a C.D. is a white

25
00:02:27,950 --> 00:02:28,910
noise or not.
