00:01
So we're looking at a regression between some different variables, and we've been given the p -values and the level of significance.
00:08
Ok, so first of all, what are we doing? we're trying to see if there's a relationship between these different variables, retail and three types of marketing.
00:21
Now, it's not reasonable to take every single sale and see why it happens.
00:28
That would just take too long.
00:29
So we take a sample and we use the sample.
00:32
Now, it's possible that maybe you take a sample of sales and you find a relationship between, say, tv and sales that doesn't actually reflect reality.
00:43
It just happens because of your sample just by random chance.
00:47
And we want to know how likely it is that something we saw happens if there were no relationship.
00:56
So the start of a test is you assume no relationship.
01:01
Relationship so that each type of marketing has no impact on sales.
01:07
Then you find the probability of our sample outcome if, and we call this the null hypothesis, if the null hypothesis is true.
01:26
So you start out by saying, okay, i think there's no relationship between tv and sales.
01:31
Then you take your sample, you look at the relationship that your sample shows and say, okay, what is the probability of this happening just by chance...