31/10/2018

# 留学申请文书：因变量如何因自变量的变化而变化 R平方是确定系数，表示两个不同级数变量之间的变化。它说明因变量如何因自变量的变化而变化(Montgomery et al.， 2012)。R方越高表示变异程度越低，R方越低表示变异程度越高。表中体重、身高、腹部、前臂、手腕的R平方分别为37.5%、0.1%、66.2%、13.1%、12.0%。腹部R平方为66.2%，高于其他表明腹部良好符合回归曲线的预测因子。调整后的平方表示变量的值，这些变量实际上导致因变量的变化。如果调整后的R平方更大，这意味着变量的数量会对因变量的值产生影响(Seber和Lee, 2012)。

R Square is the coefficient of determination which represents the variation among the variables of two different series. It states that how the dependent variable varies due to variation in independent variable (Montgomery et al., 2012). Higher the R square indicates low degree of variation and lower the R square indicate high degree of variation. In above table the R square of Weight, Height, Abdomen, Forearm, and Wrist are 37.5%, 0.1%, 66.2%, 13.1%, and 12.0% respectively. The R square of Abdomen is 66.2% which is higher than other predictors which state that the abdomen good fit to the regression line. Adjusted R square states about the value of the variables which actually cause variation in dependent variables. If the adjusted R square is higher which means the most number of variables bring effect on the value of the dependent variable and vies a verse (Seber and Lee, 2012).

As per the result found in above table, the adjusted R square for Abdomen (66%) is greater than other available predictors such as Weight (37%), Height (00%), Forearm (12.7%), and Wrist (11.7) Standard error shows the difference between the predicted value and actual value. It states that how accurately the sample represent the population. If the standard error is high, which represents that sample is not able to predict the value. Whereas, lower the standard error states that the sample is able to predict the value accurately (Sharpe et al, 2015). On the basis of analysis in above table, it can be said that the standard error with Abdomen (4.87748) is lesser Weight (6.62902), Height (8.38278), Forearm (7.81874), and Wrist (7.86575).