Binomial formula: the number of results within the specified number of tests; It is often used to indicate the success rate or failure rate of test results, for example, the number of defective products in a batch of upcoming products or the number of upcoming customers of a specific type.
Cauchy: the long extension from the center to both sides; Cauchy is usually used to simulate large-scale divergent data, which are distributed near the center of the average; Cauchy distribution looks like normal distribution, but it deviates a lot.
X distribution: when the independent variables of standard normal distribution are summed by n squares, the result of X distribution will be square; It is often used in statistical experiments.
Constant distribution: no random number is generated, and the constant value will not change; In the early stage of establishing a model, it is often used to reduce the influence of random factors or to represent the same number and quantity that have been determined.
Empirical distribution: For everyone, if they are familiar with the probability of events, users often make or define specific distribution types themselves.
Erlang: Frequency is mainly based on permutation theory, which represents the number of services in various activities and is used for telephone communication modeling.
Exponential distribution: Exponential distribution is the most commonly used in industrial and commercial services. It is mainly used to define the time interval of events, such as the time interval of customers shopping in the supermarket, the period of equipment update and maintenance, etc. It is also used for the average time of telephone conversation and the number of maintenance in a certain period of time.
Extreme value 1A: describes the maximum distribution range of various types of examples. The maximum value is often used in the parameters of astronomy, human life, radiation system, material strength, flood and earthquake analysis, rainfall prediction and other system models.
Extreme value 1B: describes the minimum distribution range of various types of examples. The minimum value is usually used for system model parameters, such as astronomy, human life, radiation system, material strength, flood and earthquake analysis and rainfall prediction.
Gamma: usually used to indicate the time required to complete a task. When the parameter value of this distribution is between 0 and 1, it is similar to a decreasing exponential distribution curve. If the parameter value is greater than 1, the distribution inclines from the peak to the minimum like a pendulum.
Geometry: in a series of independent Bernoulli experiments with a certain success rate, the number of failed events before the success of the first experiment is output. It is usually used to indicate the number of products detected before the first defective product is detected, the number of entities with random size or the number of entities required in the order.
Super-information: Super-information distribution is usually used in telephone communication and queuing theory.
Inverse Gaussin: usually used to simulate the diffusion process and boundary conditions of Brownian motion; It can also simulate the distribution of specific scale, reliability, validity period and maintenance time in the total.
Inverse Weibull: In general, the distribution is certain, but when it reaches the extreme value, the data has great deviation; This distribution is used to describe several effective processes in life distribution; It is also used to fit the extreme abnormal data of the deviation area on one side of the vertex.
Johnson SB: This distribution is a transformation of normal distribution. Johnson distribution is used to describe the non-normal process in the process of quality control, and then it can be transformed into normal distribution for standard inspection.
Johnson SU: like Johnson SB, this division is transformed from normal distribution and can also be used to describe the non-normal process in quality control. In addition, this can be used to replace the well-known unstable Pearson IV distribution, and its range of values is quite reliable.
Laplace (exponential distribution): this distribution has a sharp vertex in the middle to distinguish it from normal distribution; Laplacian distribution can be used to describe two independent distributions with the same index. Often used for error analysis.
Logarithm: Logarithm distribution can be used to describe the kind of a sample; That is, how many different types can there be in a given sample? For example, the distribution has been used for the number of people with certain characteristics in the population sucked by a mosquito, or the quantity of a certain type of goods in a set of inventory.
Logistic (mathematical distribution): The mathematical distribution is very similar to the normal distribution, but there is also a big deviation. The function of mathematical distribution is mainly used in the development mode of some problems; For example, population problems, business interests, business failures and so on.
Logarithmic Logistic: when the parameter S= 1, it is exponential distribution; When the parameter s
Lognormal (standard logarithm): this distribution is often used to describe the time required to carry out an activity (especially when there are multiple affiliated activities), the interval between activities failure or the duration of manual activities; It is also widely used to protect other commercial property insurance, such as the evaluation of stock return rate or housing investment return rate.
Negative binomial distribution: negative binomial distribution is used to describe the number of experiments that failed before the first event was successful; P stands for the possibility of success.
Normal distribution: it is the famous gaussian curve or pendulum curve; When the event is caused by objective factors rather than human factors, it is most widely used; For example, it describes the sum distribution or error distribution of many numbers.
Pareto (negative exponential distribution): defined as an exponential distribution opposite to the exponential distribution, with jumping points on the left and exponential extension lines on the right; This distribution is often used to simulate many empirical phenomena with very long extension curves, such as the income distribution of a society, the size of urban population, the emergence of natural resources, the fluctuation of stock prices, the size of companies, the brightness of comets, and a series of traffic jams.
Pearson v-type: Pearson v-type distribution is usually used to describe the time required to complete certain tasks; From the distribution density, it looks like a lognormal shape, but there is a big pole when x is close to zero.
Pearson type VI: Pearson type V distribution is usually used to describe the time required to complete certain tasks; On the left side of zero, the distribution is continuous and definite; The distribution to the right of zero is uncertain.
Poisson (Poisson distribution): Poisson distribution is mainly the ratio of simulated events; For example, the number of calls per minute, the number of typos per page or the number of events in the system within a certain period of time. Note that in queuing theory, the ratio of event arrival is usually defined as Poisson arrival per unit time, and this distribution principle is similar to exponential distribution.
Power function: both sides of the function exist, and the contained value cannot be negative. Uniform distribution is a special case of function distribution.
Rayleigh:Rayleigh often represents life (validity period) because its risk rate increases with time; For example, the life of vacuum tubes. It jumps on the left and has a long extension line.
Triangle: It is usually more suitable to represent business processes than standard distribution because it provides the most accurate initial evaluation of actual values. It is often used when only three feature information (maximum value, minimum value and maximum possible average value) are known during processing.
Uniform distribution (integer or constant): Uniform distribution (integer or real number) is used to describe that all values are possible within a specific range of values; If there is little information about the task, it is usually used to describe the duration of the task activity.
Weber distribution: Weibull is mainly used to describe the product life cycle and the reliability of the project, such as the time interval of mechanical equipment damage (TBF) and the maintenance cycle (TTR).