I have forgotten about the probability and statistics I learned during my studies, and I often use it at work. I have always wanted to learn more systematically. I have suffered from lack of good teaching materials. Recently, I found an interesting statistics course ( StatQuest! ) Now I decided to stand on the shoulders of giants and comb through the basics of statistics systematically, hoping to learn to the end~~~.
The StatQuest! series of courses was created by Josh Starmer , a senior in the genetics department of the University of North Carolina at Chapel Hill , and posted on Youtube (you can’t get on the tubing in China, but my kind friends have moved to station B)了), the course covers the following parts
- 统计学基础，Statistics Fundamentals – These videos give you a general overview of statistics as well as a be a reference for statistical concepts. Topics include:
- Histograms
- What is a statistical distribution?
- And many more!!!
- 线型回归和线型模型，Linear Regression and Linear Models – These videos teach the basics relating to one of statistics most powerful tools. Linear Regression and Linear Models allow us to use continuous values, like weight or height, and categorical values, like favorite color or favorite movie, to predict a continuous value, like age.
- 逻辑回归，Logistic Regression – These videos pick up where Linear Regression and Linear Models leave off. Now, instead of predicting something continuous, like age, we can predict something discrete, like whether or not someone will enjoy the 1990 theatrical bust Troll 2.
- 机器学习，Machine Learning – Linear Models and Logistic Regression are just the tips of the machine learning iceberg. There’s tons more to learn, and this playlist will help you trough it all, one step at a time.
- NGS，High Throughput Sequence Analysis – If you do high-throughput sequence analysis, this playlist is for you!
- R中的统计学，Statistics in R – If you want to do any of this stuff in R, this playlist is for you, and you only. No one else is allowed to watch it.
The picture below shows Josh Starmer True Venerable~~~, did you see the guy behind? That is the amount of props he teaches. He often creates BGM in the course, which is very interesting~~~
This article describes what is a statistical distribution?
Measure the height of a group of people, and then stack the height data in different bins as shown in the figure below to get a histogram .
We will find that most heights fall between 5 and 6 feet, and there are relatively few heights shorter than 5 and higher than 6 feet;
It can be understood that when a person's height is measured randomly, his height is likely to fall between 5 and 6 feet.
When the width of the bins is reduced by half, it can be more accurately found that most heights fall between 5.25 and 5.75 feet, and there are relatively few heights shorter than 4.5 and higher than 6.5 feet;
By further reducing the bin size and measuring more people, a more accurate distribution of the population height can be obtained.
A smooth curve can be used to approximate the distribution of the histogram. The height is mainly distributed between 5 and 6 feet, and the probability of being shorter than 5 and higher than 6 feet is relatively small.
The curve is better than the histogram, and it is not limited by the width of bins. Differentiation can be used to directly calculate the probability of height falling in any interval instead of using the nearest interval to estimate.
In addition, when our time and economy are limited, we can use the mean and standard deviation of the limited data to estimate the approximate curve of the population's height .
Both the curve and the histogram belong to the statistical distribution, showing the probability distribution of the measurement. A higher value indicates that the probability of the data falling in this interval is greater, and a relatively low position indicates that the probability of the data falling in this interval is smaller.
Reference