# Get hadoop certification in 2019

by Simran Aggarwal your beauty our passion
The need for Big data Hadoop professionals has increased over the last couple of years. IT guys and welcome back to this class busy and machine learning in Python. Part in this lecture we are going to continue with our example of comparing the height of men and the height of women or drug effect and placebo effect and show you how to come up with a solution quantitatively as is typical infrequent statistics. We are going to assume that our data is normally distributed So both men and women come from a Gaussian distribution with possibly different means.
So what we do is we collect some data so we measure some people and now we have two lists of heights one for men and one for women. Next, we create what is called a test statistic. Let's call it T and it is equal to the hadoop certification mean of x minus the mean of x divided by as of p times the square root of two over n so as if P is called the pooled standard deviation. And as you can see it's the average of the two individuals standard deviations. Note that these standard deviations are the unbiased estimates. So we divide by and minus one rather than. And note that and is the number of samples collected for one group.

So if you have men and women and is if you have groups of unequal lengths you need to use a different formula for the port standard deviation and the tested testing. But we'll leave that out for simplicity. So if you think back to when we were looking at our hadoop certification estimate of the mean we said that because it was a sum of random variables it was also a random variable and we showed that its distribution is Yassine. If you look at this test that mistake that we just calculated you'll notice that it is also a function of random variables.
• So it is also a random variable. The next step is kind of hand-wavy because it can be shown that the tested stick we just calculated is not Gaussian distributed but rather the distributed you'll see the t distribution along with some more exotic distributions quite a bit when you're studying hadoop certification statistical testing and bezier methods.
• So what is the t distribution? Basically, it looks a lot like the Gaussian distribution but it has fatter tails. That means there is more probability weight on the tail end compared to the Ghasia if you recall the Gaussian drops off exponentially by x squared which is a very fast drop of the t distribution has a really complicated formula for the PTF but we won't be using it in this course.
• Notice that there is only one hadoop certification parameter here. The Greek letter new which we call the degrees of freedom new controls how wide or skinny the PDA is for our statistical test new is equal to two and minus two similar to the Gaussian the t distribution can have a mean and scale parameter but again we won't be using those in this course.
• So let's look at the formula for the test that tested again and examined how the value of T might change under certain conditions. You can see that if the mean of x and x are the same then t will be zero. So it falls right in the middle of the TPD of this would be the situation where the height of men and women is the same.

Benefits of learning Big Data Hadoop
What if the men of x and x are very different well then t will be either very large or very small. Notice that like the Gaussian distribution the distribution is also symmetric. It doesn't matter if X one is men and x is hadoop certification women or if x y is women and x to his men will still get either a positive test that mistake or the negative of that same test statistic.
• We'll examine this formula more in-depth later but for now let's move on to the next step. So similar to the problem of finding a confidence interval. We want to find the area under the curve of the t distribution. So we need CTF.
• Also, remember that there are two tails of the distribution and we need to consider the left and the right. If we are at the extreme left then the area will be very small. So we'll get a probably close to hadoop certification
•  zero if we are at the extreme right then the area will be very close to.
• So for a significance level of percent then we want to be either at the bottom. percent of the area on the left or the top. percent of the area on the right. If that is the case then we can say that the difference between our two groups is statistically significant. Unfortunately, this isn't the whole story in terms of learning about statistics terminology algorithmically we are done.

We know how to show that the two groups have a significant difference. There's just more terminology and definitions that are used in statistics that we want to know about
hadoop certification so we can apply them in later lectures. Similarly it could mean that using the drug leads to worse results than a placebo or it could mean that using the drug leads to better results than a placebo we will be doing the two-sided tests or in other words our alternative hypothesis will be that the average height of men is not equal to the average height of women in the next lecture. We'll look at how to quantitatively show that the two groups are statistically significantly different.