1
00:00:01,680 --> 00:00:08,650
Before diving into how forecasting is to be done, let's cover the definition of Time, Cities and Time

2
00:00:08,650 --> 00:00:17,220
series forecasting First Time Cities is simply a series of data points where time order.

3
00:00:18,840 --> 00:00:23,760
This means that we pick one or more variables and know their values over time.

4
00:00:25,710 --> 00:00:28,260
Here are some examples of time to these data.

5
00:00:29,760 --> 00:00:34,470
The first table shows profit over last five years,

6
00:00:37,770 --> 00:00:41,440
so you can see profit from 2001 to 2005.

7
00:00:42,090 --> 00:00:47,670
And for each year we have the profit value in the second table.

8
00:00:48,720 --> 00:00:52,950
We have data noted for temperature for every five minutes.

9
00:00:53,940 --> 00:00:55,620
This is and other times it is.

10
00:00:57,240 --> 00:01:00,390
But the third example is not four times it is.

11
00:01:00,720 --> 00:01:02,040
It is cross-sectional data.

12
00:01:02,960 --> 00:01:10,080
Cross-sectional data is a set of measurements taken at one point in time or time is not considered as

13
00:01:10,080 --> 00:01:10,620
a factor.

14
00:01:12,720 --> 00:01:19,620
For example, in this third day, will we have latitude and temperature values of different cities?

15
00:01:20,190 --> 00:01:27,750
These maybe have been noted for one particular time period to time cities is simply a series of data

16
00:01:27,750 --> 00:01:28,170
points.

17
00:01:28,620 --> 00:01:39,510
With Time 90s forecasting is use of model to predict future values based on previously observed values.

18
00:01:40,800 --> 00:01:47,820
So we will look at actual historical values in the cities and try to build a model that can predict

19
00:01:47,820 --> 00:01:50,790
values based on these historical values.

20
00:01:52,590 --> 00:01:56,520
For example, this is previous year's sales performance.

21
00:01:57,420 --> 00:02:01,770
Future sales forecast can be generated for the coming time period.

22
00:02:03,360 --> 00:02:08,970
We are going to see different types of models that can help us make these predictions.