回归分析模型离差系数偏导法在建筑产品投资决策中的应用研究

Application of the Deviation Coefficient Derivative Method Based on Regression Analysis Model in Investment Decision of Building Products

建筑产品投资决策定量分析所构建的回归模型离差系数计算中,提出了运用基于OSL原理的偏导法有助于解决较为复杂的建筑市场数据线性相关问题研究.通过判断某地区建筑业增加值和全社会固定资产投资因素样本量n为10的线性相关性,计算出离差平方和最小的β_0、β_1值.经检验,在自由度8和显著性水平α为0.05时,相关系数R为0.964,大于临界值0.632;回归系数t_β为2.727,服从t(0.025,8)分布.结论表明,模型用于未来建筑市场预测的精确性,在项目投资决策中起到一定的辅助作用.

In this paper,in calculating the deviation coefficient of the regression model constructed by quantitative analysis of investment decision-making of building products,a partial derivation method based on OSL principle is proposed,which is helpful to solve the more complex linear correlation problem of construction market data.By judging the linear correlation between the added value of construction industry in a certain area and the sample size of fixed assets investment factors in the whole society,the minimum sum of squares of deviations of β0 and β1 is calculated.After inspection,the correlation coefficient R is 0.964 when the degree of freedom is 8 and the significant level α is 0.05,which is higher than the critical value 0.632.The regression coefficient tβ is 2.727,which obeys the t(0.025,8) distribution.The conclusion shows that the accuracy of the model used in future construction market prediction plays a certain auxiliary role in project investment decision-making.