In this paper, we use Monte Carlo simulation method to estimate the intra group correlation coefficient of the whole group of random sample classification design and recalculate the sample size according to the intra group correlation coefficient. Using this method, we can easily calculate the correlation coefficient and sample size and use the ODS function of SAS software to get word output. Then random samples are classified. In addition, this paper analyzes the relationship between summer heavy precipitation and spatial-temporal variation of circulation rainfall by using daily precipitation data of meteorological stations, NCEP reanalysis data, and relevant circulation index. Singular value decomposition (SVD) method is used to analyze the spatial correlation between spatial-temporal variation of summer rainfall and 500 hPa geopotential height; with the help of wavelet coherence and cross wavelet transform, the relationship between summer heavy rainfall and Western Pacific subtropical high and the relationship between East Asian summer monsoon and South China Sea summer monsoon are analyzed. Finally, the temporal and spatial variation of heavy rainfall in summer in this area is analyzed. With the development of economic globalization, the distribution of various products among countries becomes more and more frequent. In recent years, the number of imported products from China has increased, and many foreign products have entered the Chinese market. Because only in this way can people of different classes and ages know the basic characteristics and usage of the product. Standardizing English vocabulary translation of imported product description plays an important role in improving user’s product experience and increasing product sales. This paper adopts the method of random sample classification to study the temporal and spatial changes of rainfall and English vocabulary translation, so as to promote the development of imported products.