[Editor’s note]Recently, Professor Hao Xuguang from w88 casino’s International Business w88 casino published articles in Beijing Daily and Shanghai Securities News, explaining the law of large numbers in statistics and the psychology of stock investment.

There is an important law of w88 login numbers in statistics, also known asThe "Theorem of w88 login Numbers", "Law of w88 login Numbers" or "Law of Averages". In layman's terms, people have discovered through long-term practice that almost inevitable patterns often appear in w88 login numbers of repetitions of random phenomena. That is, only through the study of w88 login sample sampling can regularity be derived, and only w88 login samples can reflect the population. In fact, people are affected by the "law of small numbers", that is, when judging probability, they judge the probability of something happening based on experience and ignore the size of the sample, thinking that small samples can reflect the overall view. Kahneman and Tversky's 1974 experiment proved this.

Experimental case, there are two hospitals in a certain city, and about 100,000 babies are born in the big hospital every day45 babies, about 15 babies are born every day in the small hospital. The birth rate of boys is approximately 50%. But the exact proportion of boys born varies from day to day. Every year, hospitals count the number of days when the birth rate of boys is higher than 60%. Which hospital do you think has more? A, big hospital; B, small hospital; C, the same number. The experimental results of Kahneman and Tversky showed that when the subjects answered these three questions, 21 people thought that there were more w88 login hospitals, 21 people thought that there were more small hospitals, and 53 people thought they had the same number

The author has done many studies and the conclusions are similar to this, total1961 people answered this question, of which N(A)=417(21.3%); N(B)=601(30.6%); N(C)=943(48.1%).

The probability theory tells us that there may be more such days in small hospitals than in w88 login hospitals because the sample is larger(w88 login hospital), the closer it is to the average value of 50%, while the sample size of a small hospital is smaller, so the possibility of deviation from the average value of 50% will be greater.

ThrowIs the probability of 500 heads and 500 tails of a coin flipped 1000 times the same as the probability of 3 heads and 3 tails of a coin flipped 6 times? Nearly 40% of the subjects in the experiment thought they were the same, but in fact they were different. The former is a w88 login sample and is unlikely to deviate from the average. The latter is a small sample and more likely to deviate from the mean. People make incorrect judgments because they ignore sample size.

In fact, accidental events in small samples do not have the function of self-correction, and an extremely high score may not necessarily be balanced by a corresponding extremely low score. The rest of the sample is just a matter of chance"Dilute" rather than "neutralize" so that its average is closer to the overall average.

The reason why people make such judgment errors is that they believe that any small sample or event should have the characteristics of the entire domain.

Sample size is very important in statistics. It is a key factor in making accurate judgments about the problem. However, affected by the representativeness bias, when faced with sampling problems, people often tend to ignore the sample size if they do not think carefully, systematically and rationally. In practice, extreme results are more likely to occur in small samples than in w88 login samples.

(The author is a professor at the w88 casino of International Business, w88 casino)

Attachment: Original link:

http://bjrb.bjd.com.cn/html/2017-01/16/content_97220.htm