First, the weighted average method
The weighted average method is a simple and intuitive method to calculate the weight proportion, and its proportion in the population is calculated according to the weight of various factors or indicators (that is, subjective importance coefficient). The formula of weighted average method: wi=∑j= 1najai, wi represents the weight ratio of the i-th factor or indicator, ai represents the weight of the i-th factor or indicator, and n represents the total number of factors or indicators.
Second, the factor analysis method
Factor analysis is a method to calculate the weight proportion by using the information concentration principle of data. By extracting the hidden common factors in the data, the weight ratio is determined according to the contribution of each factor to the data variance interpretation rate. The formula of factor analysis: wi =∑j = 1λjλi, where wi represents the weight proportion of the ith common factor, λI represents the characteristic root of the ith common factor, and m represents the number of common factors.
Third, AHP analytic hierarchy process
Analytic Hierarchy Process (AHP) is a method to calculate the weight proportion by using the relative size of numbers. It expresses the relative importance of each factor or index by constructing a judgment matrix, and determines its weight ratio by solving the feature vector. The formula of AHP is wi=∑j= 1nvjvi.
Fourthly, entropy method.
1. Entropy method is a method to calculate the weight ratio by using the entropy information of data. It determines the weight ratio by calculating the information content of each factor or index in the data.
2. The formula of entropy method: wi=∑j= 1n( 1? ej) 1？ Ei and wi represent the weight ratio of the i-th factor or indicator, e i represents the entropy value of the i-th factor or indicator, and n represents the total number of factors or indicators.
3. Calculation formula of entropy value: ei=? Kj= 1∑mpijlnpij, where ei represents the entropy value of the i-th factor or index, k represents a constant, generally lnm 1, m represents the number of samples, and pij represents the normalized value of the i-th factor or index in the j-th sample.
Three applications of weight proportion
1, performance evaluation
By determining the weight proportion of different performance indicators, the performance of employees or organizations can be comprehensively evaluated and ranked.
2. Risk assessment
By determining the weight proportion of different risk factors, risks can be quantified and classified, so as to formulate corresponding risk response measures.
3, decision analysis
By determining the weight proportion of different decision-making schemes, the advantages and disadvantages of decision-making schemes can be compared and selected to achieve the optimal decision-making results.