Shulou(Shulou.com)06/02 Report--

This article introduces the knowledge of "how to use Python to explore the probability distribution of variables". In the operation of actual cases, many people will encounter such a dilemma, so let the editor lead you to learn how to deal with these situations. I hope you can read it carefully and be able to achieve something!

A brief background introduction

First of all, the normal distribution is also called Gaussian distribution.

It is named after Carl Friedrich Gauss, a mathematical genius.

Normal distribution also known as Gaussian distribution

3. The simpler the models are, the more commonly used they are because they can be well explained and understood. The normal distribution is very simple, which is why it is so commonly used.

Therefore, it is necessary to understand normal distribution. What is the probability distribution?

First of all, introduce the related concepts.

Consider a prediction model that can be a component of our data science research.

If we want to accurately predict the value of a variable, then the first thing we need to do is to understand the potential characteristics of the variable.

First of all, we need to know the possible values of the variable and whether these values are continuous or discrete. To put it simply, if we want to predict a value, then the first step is to understand that its value is 1 to 6 (discrete).

The second step is to determine the probability of each possible value (event). If a value never appears, the probability of that value is 0.

The greater the probability of an event, the more likely it is to occur.

In practice, we can repeat a large number of experiments and record the results of the corresponding output variables of the experiment.

We can divide these values into different set classes, and in each category, we record the number of times the results belong to this category. For example, we can cast 10000 times, each with 6 possible values, we can set the number of categories to 6, and then we can start counting the number of occurrences of each category.

We can draw the curve of the above results, which is the probability distribution curve. The possibility of each value of the target variable is determined by its probability distribution.

Once we know the probability distribution of variables, we can begin to estimate the probability of events, and we can even use some probability formulas. At this point, we can better understand the characteristics of variables. The probability distribution depends on some characteristics of the sample, such as average, standard deviation, skewness and kurtosis.

If you sum all the probability values, the result will be 100%.

There are many different probability distributions in the world, and the most widely used one is the normal distribution.

First encounter normal distribution

We can draw the probability distribution curve of normal distribution, and we can see that the curve is a bell-shaped curve. If the mean, modulus and median of a variable are equal, then the variable has a normal distribution.

As shown in the following figure, it is the probability distribution curve of normal distribution:

It is very important to understand and estimate the probability distribution of variables.

The distribution of the variables listed below is relatively close to the normal distribution:

The height of the crowd

Blood pressure in adults

The position of particles in propagation

Measurement error

Residual error in regression

The shoe size of the crowd

The total time it takes for employees to go home during the day

Education index

In addition, a large number of variables in life are normal variables with x% confidence, among which x

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.