(1) linear, that is, this estimator is a random variable.

(2) Unbiasedness, that is, the average or expected value E(a) of this estimator is equal to the true value A. ..

(3) It has an effective estimator, that is, this estimator has the smallest variance among all such linear unbiased estimators.

Ps: (2) Let me explain a little, that is, if y= a0+a 1*x 1+u, then unbiased means E (A0) = A0, and E (A 1) = A 1.

Multiple correlation coefficient: In fact, what you are looking for should be "correlation coefficient".

Correlation table and graph can reflect the relationship between two variables and their correlation direction, but they can't accurately show the degree of correlation between two variables.

Karl pearson, a famous statistician, designed a statistical index-correlation coefficient. Correlation coefficient is a statistical index reflecting the close correlation between variables. The correlation coefficient is calculated according to the product-difference method, which is also based on the deviation between two variables and their respective average values, and the correlation degree between the two variables is reflected by multiplying the two deviations; The linear single correlation coefficient is studied emphatically.

According to the different characteristics of related phenomena, the names of their statistical indicators are also different. For example, the statistical index reflecting the linear correlation between two variables is called correlation coefficient (the square of correlation coefficient is called judgment coefficient); The statistical indexes reflecting the curve correlation between two variables are called nonlinear correlation coefficient and nonlinear judgment coefficient. The statistical indexes reflecting multivariate linear correlation are called complex correlation coefficient and complex judgment coefficient.