Abstract: The lack of flood disaster data in many cities has led to an insufficiency of effective means to quantify flood losses in these cities. To meet the increasingly severe risk management requirements of urban flooding, there is an urgent need of implementing a quantitative assessment method for quantifying the flood losses in cities lacking disaster data. The method of "factor variation-dynamic matching-objective-driven-scenario fitting" was proposed to construct the flood loss rate functions with lack of data. In this study, based on the idea of equiproportional substitution, a variance analogy factor was constructed using multiple reference objects and multiple characteristic indicators; a dynamic analogy method was established to form a transplantation sample matrix to minimize the coefficient of variation; the water depth-loss rate fitting sequence was determined to maximize the probability of beta distribution; multiple fitting scenarios were set, and the preferred flood loss rate function was selected with the criterion of maximizing the fitting correlation coefficient. Taking Zhengzhou City as an example, the flood loss rate functions of 4 land use types were simulated. The results demonstrated that the proposed method for establishing flood loss rate function in cities lacking data was feasible; the characteristic combination indexes showed dynamic variability and the fitting effect of multiple-function combinations was optimal.