Abstract: Natural geographical factors and production input on the agricultural production has a significant influence in arid and semi-arid region. In this paper, based on the analysis of the trend of grain yield by the data of agricultural food production during 1980-2017 in the Guanzhong area of Shaanxi Province,the PCR(principal component regression simulation)equation of the agricultural production in Guanzhong area is established by the help of PCA( principal component analysis ) theory, then, on the basis of calculating the dependent variable (Y) elastic coefficient of regression equation, the performance contribution of geographical factors and production input index (Y) to grain yield of guanzhong area was analyzed and studied. The natural geographical factors include: Y₁ (farmland cultivation area),Y₂ (the main food crop farming area),Y₃ (grain stable yield farmland area),Y₄ (the effective irrigation farmland area), Y₅ (affected farmland area),Y₆(disaster farmland area); the production input factors include: Y₇ (total amount of fertilizer applied in agriculture),Y₈ (total power of agricultural machinery),Y₉ (agricultural electricity consumption),Y₁₀ (number of agricultural labor force),Y₅ (total pesticide usage),Y₁₂ (agricultural film usage). According to the calculating data of the PCR model analysis, the main results show that: the factors of independent variable whose average elastic coefficient is greater than 0 are Y₃(0.117),Y₄(0.509),Y₇(0.793), Y₈(0.091), Y₉(0.478),Y₁₀ (0.106), Y₁₂(0,002) and it had greatly promoted the development of agriculture. Less than 0 of the factors are Y₁ (-0.763), Y₂ (-0,009), Y₅ (-0.052),Y₆ (-0,062), Y₁₁(-0,001) and it were a more obvious obstacle to improve the efficiency of agricultural production. In dependent variable (Y) of the geographical factors, the index Y₃ and Y₄,Whose 75% of the elastic coefficients are greater than 0, are the main contributions to the performance of grain production in Guanzhong area. In dependent variable (Y) of the production input, the indexs Y₇,Y₈ and Y₉, Whose 79.2% of the elastic coefficients are greater than 0, are the main contributions to the performance of grain production in Guanzhong area. At the same time, it also explains the reason why grain production of Guanzhong area presents an trend of periodic change and continuous growth.

Key words: Guanzhong Area, geographical factors, principal component regression(PCR), coefficient